DataCast

Follow DataCast
Share on
Copy link to clipboard

پادکست دیتاکست پادکستی در مورد Big Data Data Mining Machine Learning وسایر اصطلاحاتی که امروزه ترند شده

Mohammad Reza Mashoufi


    • Sep 9, 2021 LATEST EPISODE
    • every other week NEW EPISODES
    • 1h 4m AVG DURATION
    • 71 EPISODES


    Search for episodes from DataCast with a specific topic:

    Latest episodes from DataCast

    Episode 71: Trusted AI with Saishruthi Swaminathan

    Play Episode Listen Later Sep 9, 2021 73:46


    Timestamps(01:59) Saishruthi talked about her upbringing, growing up in a rural town in India with no Internet connection and no computers.(05:50) Saishruthi discussed her undergraduate studying Electrical Engineering at Sri Sairam Engineering College in the early 2010s.(11:56) Saishruthi mentioned the projects and learnings during her two years working at Tata Consultancy Services as an instrumentation engineer.(15:57) Saishruthi went over her MS degree in Electrical Engineering at San Jose State University and her journey into data science.(22:20) Saishruthi shared the initial hurdles she faced transitioning back to school and assimilating to the US culture.(26:10) Saishruthi touched on her work with San Jose City on disaster management.(28:20) Saishruthi went over her job search process, eventually landing a data science position at IBM.(32:16) Saishruthi unpacked lessons learned from public speaking.(35:20) Saishruthi summarized IBM's data science and machine learning initiatives.(37:02) Saishruthi brought up various projects happening at IBM's Center for Open Source Data and AI Technologies, whose mission is to make open-source AI models dramatically easier to create, deploy, and manage in the enterprise.(39:40) Saishruthi unpacked the qualities needed to contribute to open-source projects and their role in shaping the development of ML technologies.(44:50) Saishruthi dissected examples of bias in ML, identified solutions to combat unwanted bias, and presented tools for that (as delivered in her talk titled “Digital Discrimination: Cognitive Bias in Machine Learning”).(49:12) Saishruthi shared her thoughts on the evolution of research and applications within the Trusted AI landscape.(54:07) Saishruthi discussed the core value propositions of IBM's Elyra, a set of AI-centric extensions to JupyterLab that aims to help data practitioners deal with the complexities of the model development lifecycle.(56:11) Saishruthi briefly shared the challenges with developing Coursera courses on data visualization with Python and with R.(01:00:47) Saishruthi went over her passion for movements such as Women In Tech and Girls Who Code.(01:03:27) Saishruthi shared details about her initiative to bring education to rural children.(01:06:36) Closing segment.Saishruthi's Contact InfoTwitterLinkedInMediumGitHubCourseraMentioned ContentTalks“Digital Discrimination: Cognitive Bias in Machine Learning” (All Things Open 2020)ProjectsAI Fairness 360AI Explainability 360Adversarial Robustness ToolkitModel Asset ExchangeData Asset ExchangeElyraCoursesData Visualization with PythonData Visualization with RAbout the showDatacast features long-form, in-depth conversations with practitioners and researchers in the data community to walk through their professional journeys and unpack the lessons learned along the way. I invite guests coming from a wide range of career paths — from scientists and analysts to founders and investors — to analyze the case for using data in the real world and extract their mental models (“the WHY and the HOW”) behind their pursuits. Hopefully, these conversations can serve as valuable tools for early-stage data professionals as they navigate their own careers in the exciting data universe.Datacast is produced and edited by James Le. Get in touch with feedback or guest suggestions by emailing khanhle.1013@gmail.com.Subscribe by searching for Datacast wherever you get podcasts or click one of the links below:Listen on SpotifyListen on Apple PodcastsListen on Google PodcastsIf you're new, see the podcast homepage for the most recent episodes to listen to, or browse the full guest list.

    Episode 70: Machine Learning Testing with Mohamed Elgendy

    Play Episode Listen Later Aug 30, 2021 62:14


    Timestamps(01:44) Mohamed described his interest growing up in Egypt and studying Biomedical Engineering at Cairo University in the early 2000s.(04:22) Mohamed commented on his experience moving to the US to pursue an MBA degree and working in various software engineering roles.(07:35) Mohamed shared his experience authoring two books: (1) 3D Business Analyst: The Ultimate Hands-On Guide to Mastering Business Analysis and (2) Business Analysis for Beginners: Jump-Start Your BA Career in 4 Weeks.(13:19) Mohamed discussed his move to the Bay Area for a Senior Engineering Manager role at Twilio, managing and shipping a series of communication API products using Machine and Deep Learning.(17:39) Mohamed dissected engineering challenges building ML systems at Amazon, alongside key leadership lessons he acquired from managing Amazon's Kindle mobile and ML engineering teams.(20:50) Mohamed shared his insider perspective on Amazon's practices of customer obsession, working backward, and disagree-to-commit.(24:52) Mohamed mentioned the benefits of teaching a computer vision course for engineers at Amazon's internal Machine Learning university.(28:33) Mohamed went over the engineering (hardware + software) and ML challenges associated with building a proprietary threat detection platform at Synapse Tech Corporation (where he was the Head of Engineering).(32:03) Mohamed shared concrete technical challenges with building an ML system that performs inference on edge devices.(37:03) Mohamed revealed specific data labeling challenges while building the ML system at Synapse.(39:57) Mohamed went over his one year as the VP of Engineering for the AI Platform at Rakuten, when he incubated the idea for Kolena.(42:52) Mohamed explained the current state of ML testing infrastructure and unpacked his current project Kolena, a rigorous ML QA platform that lets users take control of their ML testing.(49:07) Mohamed has been collaborating with a few institutions, podcasters, and ML influencers to raise awareness of the importance of ML testing and different approaches to tackle the problem.(50:12) Mohamed touched on his side hustles working with Intel in autonomous drones and teaching content with Udacity's AI Nanodegree programs.(53:07) Mohamed dissected his project Mowgly, an educational platform with tracks curated by industry experts to guide users to master specific topics.(54:58) Mohamed described his experience authoring a book with Manning in 2020 called “Deep Learning For Vision Systems.”(58:51) Closing segment.Mohamed's Contact InfoLinkedInTwitterWebsiteYouTubeGitHubKolenaMentioned ContentPeopleAndrew Trask (Leader at OpenMined, Senior Research Scientist at DeepMind, Ph.D. Student at the University of Oxford)Francois Chollet (Senior Software Engineer at Google, Creator of Keras)Lex Fridman (Host of the popular Lex Fridman Podcast, AI Researcher working on autonomous vehicles and human-robot interaction at MIT)Books“Mindset” (by Carol Dweck)“Outliers” (by Malcolm Gladwell)NotesMy conversation with Mohamed was recorded back in March 2021. Here are some updates that Mohamed shared with me since then:Kolena is an ML testing and validation platform that enables teams to implement testing best practices to rigorously test their models' behavior and ship high-quality ML products much faster.Mohamed and his team have signed a couple of big enterprise customers and raised a large seed round from top-tier investors and almost every industry leader in the AI space. These were strong signals that Kolena is solving a very important problem!Mohamed's first impression on the market is: the ML market is hungry for a reliable testing platform for models. Kolena has quite of a waitlist and plans to launch early next year.About the showDatacast features long-form, in-depth conversations with practitioners and researchers in the data community to walk through their professional journeys and unpack the lessons learned along the way. I invite guests coming from a wide range of career paths — from scientists and analysts to founders and investors — to analyze the case for using data in the real world and extract their mental models (“the WHY and the HOW”) behind their pursuits. Hopefully, these conversations can serve as valuable tools for early-stage data professionals as they navigate their own careers in the exciting data universe.Datacast is produced and edited by James Le. Get in touch with feedback or guest suggestions by emailing khanhle.1013@gmail.com.Subscribe by searching for Datacast wherever you get podcasts or click one of the links below:Listen on SpotifyListen on Apple PodcastsListen on Google PodcastsIf you're new, see the podcast homepage for the most recent episodes to listen to, or browse the full guest list.

    Episode 69: DataPrepOps, Active Learning, and Team Management with Jennifer Prendki

    Play Episode Listen Later Aug 16, 2021 88:14


    Show Notes(01:46) Jennifer shared her formative experiences growing up in France and wanting to be a physicist.(03:04) Jennifer unpacked the evolution of her academic journey in France — getting Physics degrees at Louis Pasteur University, Paris-Sud University, and Sorbonne University.(06:44) Jennifer mentioned her time as a Postdoctoral Researcher in Neutrino Physics at Duke University, where her research group lacked the funding to carry on scientific projects.(09:35) Jennifer discussed her transition from academia to industry, working as a Quantitative Research Scientist at Quantlab Financial in Houston.(13:31) Jennifer went over her move to the Bay Area, working for YuMe and Ayasdi — growing and managing early-stage data science teams at both places.(19:19) Jennifer recalled her foray into becoming a Senior Data Science Manager of the Search team at Walmart Labs. She managed the Metrics-Measurements-Insights team and the Store-Search team.(23:59) Jennifer shared the business anecdote that made her obsessed with measuring the ROI of data science.(28:46) Jennifer reflected on the opportunity to give conference talks and become a thought leader in the data science community (watch her first industry talk, “Review Analysis: An Approach to Leveraging User-Generated Content in the Context of Retail” at MLconf 2016).(31:10) Jennifer unpacked her interest in active learning and outlined existing challenges of making active learning performant in real-world ML systems.(36:58) After 1.5 years with Walmart Labs, Jennifer became the Chief Data Scientist at Atlassian. She shared the tactics to grow the Search & Smarts team of scientists and engineers from 3 to 17 people in less than 6 months across 3 locations.(40:31) Jennifer discussed the organizational and operational challenges with making ML useful in enterprises and the importance of data preparation in the modern ML stack.(47:24) Jennifer elaborated on the topic of “Agile for Data Science Teams,” which discusses that organizations that invest in ML but do not get the organizational side of things right will fail.(53:09) Jennifer went over her decision to accept a VP of Machine Learning role at Figure Eight, then a frontier startup that offers enterprise-grade labeling solutions to ML teams.(57:56) Jennifer went over the inception of her startup Alectio, whose mission is to help companies do ML more efficiently with fewer data and help the world do ML more sustainably by reducing the industry's carbon footprint.(01:04:32) Jennifer unpacked her 4-part blog series about responsible AI that calls out the need to fight bias, increase accessibility, and create more opportunities in AI.(01:09:06) Jennifer discussed the hurdles she had to jump through to find early adopters of Alectio.(01:11:03) Jennifer emphasized the valuable lessons learned to attract the right people who are excited about Alectio's mission.(01:14:38) Jennifer cautioned the danger of taking advice without thinking through how it can be applied to one's career.(01:17:09) Jennifer condensed her decade of experience navigating the tech industry as a woman into concrete advice.(01:19:19) Closing segment.Jennifer's Contact InfoLinkedInTwitterMediumAlectio's ResourcesWebsiteTwitterLinkedInWhat Is Alectio? (Video)Is Big Data Dragging Us Towards Another AI Winter? (Article)Mentioned ContentTalksThe Day Big Data Died (Oct 2020 @ Interop Digital)The Importance of Ethics in Data Science (Keynote @ Women in Analytics Conference 2019)Introduction to Active Learning (ODSC London 2018)Agile for Data Science Teams (Strata Data Conf — New York 2018)Big Data and the Advent of Data Mixology (Interop ITX — The Future of Data Summit 2017)The Limitations of Big Data In Predictive Analytics (DataEngConf SF 2017)Review Analysis: An Approach to Leveraging User-Generated Content in the Context of Retail (MLconf 2016)Articles1 — Women vs. The Workplace SeriesGender Discrimination (Oct 2015)Why Leading By Example Matters (Jan 2017)Data Scientist: the SexISTiest Job of the 21st Century? (Feb 2017)The Role of Motherhood in Gender Discrimination (March 2017)The Biggest Challenges of the Female Manager (May 2017)Parity in the Workplace: Why We Are Not There Yet (Dec 2017)The Pyramid of Needs of Professional Women (Dec 2017)2 — Management SeriesThe Secrets to Successfully Managing an Underperformer (June 2017)The Top Secrets to Managing a Rockstar (July 2017)The Real Cost of Hiring Over-Qualified Candidates in Technology (March 2018)Team Culture (May 2018)3 — Responsible AI SeriesHow We Got Responsible AI All Wrong (Part 1)Impact, Bias, and Sustainability in AI (Part 2)Increasing Accessibility to AI (Part 3)Creating More Opportunities in AI (Part 4)Book“Managing Up” (by Rosanne Badowski and Roger Gittines)NotesJennifer told me that Alectio is about to launch a community version that people will be able to compete to get the best model with the minimum amount of data this fall. Be sure to check out their blog and follow them on LinkedIn!About the showDatacast features long-form, in-depth conversations with practitioners and researchers in the data community to walk through their professional journeys and unpack the lessons learned along the way. I invite guests coming from a wide range of career paths — from scientists and analysts to founders and investors — to analyze the case for using data in the real world and extract their mental models (“the WHY and the HOW”) behind their pursuits. Hopefully, these conversations can serve as valuable tools for early-stage data professionals as they navigate their own careers in the exciting data universe.Datacast is produced and edited by James Le. Get in touch with feedback or guest suggestions by emailing khanhle.1013@gmail.com.Subscribe by searching for Datacast wherever you get podcasts or click one of the links below:Listen on SpotifyListen on Apple PodcastsListen on Google PodcastsIf you're new, see the podcast homepage for the most recent episodes to listen to, or browse the full guest list.

    Episode 68: Threat Intelligence, Venture Stamina, and Data Investing with Sarah Catanzaro

    Play Episode Listen Later Jul 14, 2021 76:06


    Show Notes(01:48) Sarah talked about the formative experiences of her upbringing: growing up interested in the natural sciences and switching focus on terrorism analysis after experiencing the 9/11 tragedy with her own eyes.(04:07) Sarah discussed her experience studying International Security Studies at Stanford and working at the Center for International Security and Cooperation.(07:15) Sarah recalled her first job out of college as a Program Director at the Center for Advanced Defense Studies — collaborating with academic researchers to develop computational approaches that counter terrorism and piracy.(09:48) Sarah went over her time as a cyber-intelligence analyst at Cyveillance, which provided threat intelligence services to enterprises worldwide.(12:22) Sarah walked over her time at Palantir as an embedded analyst, where she observed the struggles that many agencies had with data integration and modeling challenges.(15:26) Sarah unpacked the challenges of building out the data team and applying the data work at Mattermark.(20:15) Sarah shared her opinion on the career trajectory for data analysts and data scientists, given her experience as a manager for these roles.(23:43) Sarah shared the power of having a peer group and building a team culture that she was proud of at Mattermark.(26:41) Sarah joined Canvas Ventures as a Data Partner in 2016 and shared her motivation for getting into venture capital.(29:47) Sarah revealed the secret sauce to succeed in venture — stamina.(32:00) Sarah has been an investor at Amplify Partners since 2017 and shared what attracted her about the firm's investment thesis and the team.(35:28) Sarah walked through the framework she used to prove her value upfront as the new investor at Amplify.(38:35) Sarah shared the details behind her investment on the Series A round for OctoML, a Seattle-based startup that leverages Apache TVM to enable their clients to simply, securely, and efficiently deploy any model on any hardware backend.(44:39) Sarah dissected her investment on the seed round for Einblick, a Boston-based startup that builds a visual computing platform for BI and analytics use cases.(48:45) Sarah mentioned the key factors inspiring her investment in the seed round for Metaphor Data, a meta-data platform that grew out of the DataHub open-source project developed at LinkedIn.(53:57) Sarah discussed what triggered her investment in the Series A round for Runway, a New York-based team building the next-generation creative toolkit powered by machine learning.(58:36) Sarah unpacked the advice she has been giving her portfolio companies in hiring decisions and expanding their founding team (and advice they should ignore).(01:01:29) Sarah went over the process of curating her weekly newsletter called Projects To Know (active since 2019).(01:05:00) Sarah predicted the 3 trends in the data ecosystem that will have a disproportionately huge impact in the future.(01:11:15) Closing segment.Sarah's Contact InfoAmplify PageTwitterLinkedInMediumAmplify Partners' ResourcesWebsiteTeamPortfolioBlogMentioned ContentBlog PostsOur Investment in OctoMLAnnouncing Our Investment in EinblickOur Investment in Metaphor DataOur Series A Investment in RunwayPeopleSunil Dhaliwal (General Partner at Amplify Partners)Mike Dauber (General Partner at Amplify Partners)Lenny Pruss (General Partner at Amplify Partners)Mike Volpi (Co-Founder and Partner at Index Ventures)Gary Little (Co-Founder and General Partner at Canvas Ventures)Book“Zen and the Art of Motorcycle Maintenance” (by Robert Pirsig)New UpdatesSince the podcast was recorded, Sarah has been keeping her stamina high!Her investments in Hex (data workspace for teams) and Meroxa (real-time data platform) have been made public.She has also spoken at various panels, including SIGMOD, REWORK, University of Chicago, and Utah Nerd Nights.Be sure to follow @sarahcat21 on Twitter to subscribe to her brain on the intersection of data, VC, and startups!

    Episode 67: Model Observability, AI Bias, and ML Infrastructure Ecosystem with Aparna Dhinakaran

    Play Episode Listen Later Jun 28, 2021 48:11


    Show Notes(01:39) Aparna talked about her Bachelor's degree in Electrical Engineering and Computer Science at UC Berkeley.(02:50) Aparna shared her undergraduate research experience at the Energy and Sustainable Technologies lab.(04:34) Aparna discussed valuable lessons learned from her industry internships at TubeMogul and compared the objective with that of a research environment.(08:26) Aparna then joined Uber as a software engineer on the Marketplace Forecasting team, where she led the development of Uber's first model lifecycle management system for running ML model computations at scale to power Uber's dynamic pricing algorithms.(12:40) Aparna talked about how she became interested in model monitoring while Uber's model store.(17:29) Aparna discussed her decision to join the Ph.D. program in Computer Vision at Cornell University, specifically about bias in model, after spending 3 years at Uber.(23:40) Aparna shared the backstory behind co-founding MonitorML with her brother Eswar and going through the 2019 summer batch of Y-Combinator.(26:47) Aparna discussed the acquisition of MonitorML by Arize AI, where she's currently the Chief Product Officer.(28:41) Aparna unpacked the key insights in her ongoing ML Observability blog series, which argues that model observability is the foundational platform that empowers teams to continually deliver and improve results from the lab to production.(33:17) Aparna shared her verdict for the ML tooling ecosystem in the upcoming years from her in-depth exploration of ML infrastructure tools covering data preparation, model building, model validation, and model serving.(37:01) Aparna briefly shared the challenges encountered to get the first cohort of customers for Arize.(39:23) Aparna went over valuable lessons to attract the right people who are excited about Arize's mission.(41:04) Aparna shared her advice for founders who are in the process of finding the right investors for their companies.(42:24) Aparna reasoned how participating in The Amazing Race was similar to running a startup.(44:59) Closing segment.Aparna's Contact InfoTwitterLinkedInMediumForbes ColumnWebsiteGithubGoogle ScholarArize's ResourcesWebsiteMediumLinkedInTwitterMentioned ContentBlog PostsML Infrastructure Tools for Data Preparation (May 2020)ML Infrastructure Tools for Model Building (May 2020)ML Infrastructure Tools for Production (Part 1) (May 2020)ML Infrastructure Tools for Production (Part 2) (Sep 2020)ML Infrastructure Tools — ML Observability (Feb 2021)The Model's Shipped — What Could Possibly Go Wrong? (Feb 2021)PeopleRediet Abebe (Assistant Professor of Computer Science at UC Berkeley and Junior Fellow at the Harvard Society of Fellows)Timnit Gebru (Founder of Black in AI, Ex-Research Scientist at Google)Serge Belongie (Professor of Computer Science at Cornell and Aparna's past Ph.D. advisor)Solon Barocas (Principal Researcher at Microsoft Research and Adjunct Assistant Professor of Information Science at Cornell)Manish Raghavan (Ph.D. candidate in the Computer Science department at Cornell)Kate Crawford (Principal Researcher at Microsoft Research and Co-founder/Director of research at NYU's AI Now Institute)Book“The Hard Thing About The Hard Things” (by Ben Horowitz)New UpdatesSince the podcast was recorded, a lot has happened at Arize AI!Aparna has continued writing the ML observability series: The Playbook to Monitor Your Model's Performance in Production (March 2021) and Beyond Monitoring: The Rise of Observability (May 2021).Arize has been recognized in Forbes's AI 50 2021: Most Promising AI Companies.Aparna has also contributed to Forbes various articles: from the Chronicles of AI Ethics and Q&A with Ethics researchers, to a list of Women in AI to watch and emerging ML tooling categories.About The ShowDatacast features long-form, in-depth conversations with practitioners and researchers in the data community to walk through their professional journeys and unpack the lessons learned along the way. I invite guests coming from a wide range of career paths — from scientists and analysts to founders and investors — to analyze the case for using data in the real world and extract their mental models (“the WHY”) behind their pursuits. Hopefully, these conversations can serve as valuable tools for early-stage data professionals as they navigate their own careers in the exciting data universe.Datacast is produced and edited by James Le. Get in touch with feedback or guest suggestions by emailing khanhle.1013@gmail.com.Subscribe by searching for Datacast wherever you get podcasts or click one of the links below:Listen on SpotifyListen on Apple PodcastsListen on Google PodcastsIf you're new, see the podcast homepage for the most recent episodes to listen to, or browse the full guest list.

    Episode 66: Monitoring Models in Production with Emeli Dral

    Play Episode Listen Later Jun 9, 2021 46:16


    Show Notes(02:07) Emeli shared her educational background getting degrees in Applied Mathematics and Informatics from the Peoples' Friendship University of Russia in the early 2010s.(04:24) Emeli went over her experience getting a Master's Degree at Yandex School of Data Analysis.(07:06) Emeli reflected on lessons learned from her first job out of university working as a Software Developer at Rambler, one of the biggest Russian web portals.(09:33) Emeli walked over her first year as a Data Scientist developing e-commerce recommendation systems at Yandex.(13:38) Emeli discussed core projects accomplished as the Chief Data Scientist at Yandex Data Factory, Yandex's end-to-end data platform.(17:52) Emeli shared her learnings transitioning from an IC to a manager role.(19:21) Emeli mentioned key components of success for industrial AI, given her time as the co-founder and Chief Data Scientist at Mechanica AI.(22:40) Emeli dissected the makings of her Coursera specializations — “Machine Learning and Data Analysis” and “Big Data Essentials.”(26:14) Emeli discussed her teaching activities at Moscow Institute of Physics and Technology, Yandex School of Data Analysis, Harbour.Space, and Graduate School of Management — St. Petersburg State University.(30:12) Emeli shared the story behind the founding of Evidently AI, which is building a human interface to machine learning, so that companies can trust, monitor, and improve the performance of their AI solutions.(32:32) Emeli explained the concept of model monitoring and exposed the monitoring gap in the enterprise (read Part 1 and Part 2 of the Monitoring series).(34:13) Emeli looked at possible data quality and integrity issues while proposing how to track them (read Part 3, Part 4, and Part 5 of the Monitoring series).(36:47) Emeli revealed the pros and cons of building an open-source product.(39:13) Emeli talked about prioritizing product roadmap for Evidently AI.(41:24) Emeli described the data community in Moscow.(42:03) Closing segment.Emeli's Contact InfoLinkedInTwitterCourseraGitHubMediumEvidently AI's ResourcesWebsiteTwitterLinkedInGitHubDocumentationMentioned ContentBlog PostsML Monitoring, Part 1: What Is It and How It Differs? (Aug 2020)ML Monitoring, Part 2: Who Should Care and What We Are Missing? (Aug 2020)ML Monitoring, Part 3: What Can Go Wrong With Your Data? (Sep 2020)ML Monitoring, Part 4: How To Track Data Quality and Data Integrity? (Oct 2020)ML Monitoring, Part 5: Why Should You Care About Data And Concept Drift? (Nov 2020)ML Monitoring, Part 6: Can You Build a Machine Learning Model to Monitor Another Model? (April 2021)Courses“Machine Learning and Data Analysis”“Big Data Essentials”PeopleYann LeCun (Professor at NYU, Chief AI Scientist at Facebook)Tomas Mikolov (the creator of Word2Vec, ex-scientist at Google and Facebook)Andrew Ng (Professor at Stanford, Co-Founder of Google Brain, Coursera, and Landing AI, Ex-Chief Scientist at Baidu)Book“The Elements of Statistical Learning” (by Trevor Hastie, Robert Tibshirani, and Jerome Friedman)New UpdatesSince the podcast was recorded, a lot has happened at Evidently! You can use this open-source tool (https://github.com/evidentlyai/evidently) to generate a variety of interactive reports on the ML model performance and integrate it into your pipelines using JSON profiles.This monitoring tutorial is a great showcase of what can go wrong with your models in production and how to keep an eye on them: https://evidentlyai.com/blog/tutorial-1-model-analytics-in-production.About The ShowDatacast features long-form conversations with practitioners and researchers in the data community to walk through their professional journey and unpack the lessons learned along the way. I invite guests coming from a wide range of career paths - from scientists and analysts to founders and investors — to analyze the case for using data in the real world and extract their mental models (“the WHY”) behind their pursuits. Hopefully, these conversations can serve as valuable tools for early-stage data professionals as they navigate their own careers in the exciting data universe.Datacast is produced and edited by James Le. Get in touch with feedback or guest suggestions by emailing khanhle.1013@gmail.com.Subscribe by searching for Datacast wherever you get podcasts or click one of the links below:Listen on SpotifyListen on Apple PodcastsListen on Google PodcastsIf you're new, see the podcast homepage for the most recent episodes to listen to, or browse the full guest list.

    Episode 65: Chaos Theory, High-Frequency Trading, and Experimentations at Scale with David Sweet

    Play Episode Listen Later May 30, 2021 56:47


    Show Notes(01:59) David recalled his undergraduate experience studying Physics and Mathematics at Duke University back in the early 90s.(05:55) David reflected on his decision to pursue a Ph.D. in Physics at the University of Maryland, College Park, specializing in Nonlinear Dynamics and Chaos Theory.(10:18) David unpacked his Nature paper called “Topology in Chaotic Scattering.”(14:43) David went over his two papers on fractal dimensions in higher-dimensional chaotic scattering following his Nature publication.(21:42) David talked about his project K Desktop Environment, which provides a free, user-friendly desktop for Linux/UNIX systems (later turned into a print book with MacMillan Publishing in 2000).(24:20) David explained the premise behind his work on Andamooka, a site that supports open content.(27:24) David walked over his time as a quantitative analyst at Thales Fund Management after finishing his Ph.D.(30:50) David discussed his 4-year stint at Lehman Brothers — moving up the ladder into a Vice President role, up until Barclay's Capital acquired it.(33:24) David talked about his proudest accomplishment during the 5-year stint as a headdesk in equities trader at KCG/GETCO.(35:37) David shared war stories while working at an investment firm called Teza Technologies and co-founding Galaxy Digital Trading (specializing in cryptocurrency trading).(41:34) David unpacked key concepts covered in his guest lectures on optimization of high-frequency trading systems at NYU Stern School of Business.(44:26) David explained his career change to work as a Machine Learning Engineer at Instagram in the summer of 2019.(47:17) David briefly mentioned his transition back to a quant trader role at 3Red Partners.(48:05) David is writing a technical book with Manning called “Tuning Up,” which provides a toolbox of experimental methods that will boost the effectiveness of machine learning systems, trading strategies, infrastructure, and more.(50:48) David reflected on the benefits of his physics academic background for his quant analyst career.(52:27) Closing segment.David's Contact InfoWebsiteLinkedInTwitterMentioned ContentPublications"Topology In Chaotic Scattering" (Nature, May 1999)"Fractal Dimension of Higher-Dimensional Chaotic Repellors" (June 1999)"Fractal Basin Boundaries in Higher-Dimensional Chaotic Scattering"Book“The Elements of Statistical Learning” (by Trevor Hastie, Robert Tibshirani, and Jerome Friedman)PeopleJim Simons (Founder of Renaissance Technologies)Michael Kearns (Professor at the University of Pennsylvania, previously leading Morgan Stanley's AI Center of Excellence)Vasant Dhar (Professor at NYU Stern School of Business, Founder of SCT Capital)Tuning Up — From A/B testing to Bayesian optimizationManning's permanent 40% discount code (good for all Manning products in all formats) for Datacast listeners: poddcast19.You can refer to this link: http://mng.bz/4MAR.Here are two free eBook codes to get copies of Tuning Up for two lucky Datacast listeners: tngdtcr-AB2C and tngdtcr-6D43You can refer to this link: http://mng.bz/G6Bq.

    Episode 64: Improving Access to High-Quality Data with Fabiana Clemente

    Play Episode Listen Later May 18, 2021 56:21


    Show Notes(02:06) Fabiana talked about her Bachelor’s degree in Applied Mathematics from the University of Lisbon in the early 2010s.(04:18) Fabiana shared lessons learned from her first job out of college as a Siebel and BI Developer at Novabase.(05:13) Fabiana discussed unique challenges while working as an IoT Solutions Architect at Vodafone.(09:56) Fabiana mentioned projects she contributed to as a Data Scientist at startups such as ODYSAI and Habit Analytics.(12:44) Fabiana talked about the two Master’s degrees she got while working in the industry (Applied Econometrics from Lisbon School of Economics and Management and Business Intelligence from NOVA IMS Information Management School).(14:41) Fabiana distinguished the difference between data science and business intelligence.(18:01) Fabiana shared the founding story of YData, the first data-centric platform with synthetic data, whose she is currently the Chief Data Officer.(21:32) Fabiana discussed different techniques to generate synthetic data, including oversampling, Bayesian Networks, and generative models.(24:01) Fabiana unpacked the key insights in her blog series on generating synthetic tabular data.(29:40) Fabiana summarized novel design and optimization techniques to cope with the challenges of training GAN models.(33:44) Fabiana brought up the benefits of using Differential Privacy as a complement to synthetic data generation.(38:07) Fabiana unpacked her post “The Cost of Poor Data Quality,” — where she defined data quality as data measures based on factors such as accuracy, completeness, consistency, reliability, and above all, whether it is up to date.(42:11) Fabiana explained the important role that data quality plays in ensuring model explainability.(44:57) Fabiana reasoned about YData’s decision to pursue the open-source strategy.(47:47) Fabiana discussed her podcast called “When Machine Learning Meets Privacy” in collaboration with the MLOps Slack community.(49:14) Fabiana briefly shared the challenges encountered to get the first cohort of customers for YData.(50:12) Fabiana went over valuable lessons to attract the right people who are excited about YData’s mission.(51:52) Fabiana shared her take on the data community in Lisbon and her effort to inspire more women to join the tech industry.(53:47) Closing segment.Fabiana’s Contact InfoLinkedInMediumTwitterYData’s ResourcesWebsiteGithubLinkedInTwitterAngelListSynthetic Data CommunityMentioned ContentBlog PostsSynthetic Data: The Future Standard for Data Science Development (April 2020)Generating Synthetic Tabular Data with GANs — Part 1 (May 2020)Generating Synthetic Tabular Data with GANs — Part 2 (May 2020)What Is Differential Privacy? (May 2020)What Is Going On With My GAN? (July 2020)How To Generate Synthetic Tabular Data? Wasserstein Loss for GANs (Sep 2020)The Cost of Poor Data Quality (Sep 2020)How Can I Explain My ML Models To The Business? (Oct 2020)Synthetic Time-Series Data: A GAN Approach (Jan 2021)Podcast“When Machine Learning Meets Privacy”PeopleJean-Francois Rajotte (Resident Data Scientist at the University of British Columbia)Sumit Mukherjee (Associate Professor of Statistics at Columbia University)Andrew Trask (Leader at OpenMined, Research Scientist at DeepMind, Ph.D. Student at the University of Oxford)Théo Ryffel (Co-Founder of Arkhn, Ph.D. Student at ENS and INRIA, Leader at OpenMined)Recent Announcements/ArticlesPartnerships with UbiOps and AlgorithmiaThe rise of DataPrepOps (March 2021)From model-centric to data-centric (March 2021)

    Episode 63: Real-World Transfer Learning with Azin Asgarian

    Play Episode Listen Later May 6, 2021 66:00


    Show Notes(02:06) Azin described her childhood growing up in Iran and going to a girls-only high school in Tehran designed specifically for extraordinary talents.(05:08) Azin went over her undergraduate experience studying Computer Science at the University of Tehran.(10:41) Azin shared her academic experience getting a Computer Science MS degree at the University of Toronto, supervised by Babak Taati and David Fleet.(14:07) Azin talked about her teaching assistant experience for a variety of CS courses at Toronto.(15:54) Azin briefly discussed her 2017 report titled “Barriers to Adoption of Information Technology in Healthcare,” which takes a system thinking perspective to identify barriers to the application of IT in healthcare and outline the solutions.(19:35) Azin unpacked her MS thesis called “Subspace Selection to Suppress Confounding Source Domain Information in AAM Transfer Learning,” which explores transfer learning in the context of facial analysis.(28:48) Azin discussed her work as a research assistant at the Toronto Rehabilitation Institute, working on a research project that addressed algorithmic biases in facial detection technology for older adults with dementia.(33:02) Azin has been an Applied Research Scientist at Georgian since 2018, a venture capital firm in Canada that focuses on investing in companies operating in the IT sectors.(38:20) Azin shared the details of her initial Georgian project to develop a robust and accurate injury prediction model using a hybrid instance-based transfer learning method.(42:12) Azin unpacked her Medium blog post discussing transfer learning in-depth (problems, approaches, and applications).(48:18) Azin explained how transfer learning could address the widespread “cold-start” problem in the industry.(49:50) Azin shared the challenges of working on a fintech platform with a team of engineers at Georgian on various areas such as supervised learning, explainability, and representation learning.(51:46) Azin went over her project with Tractable AI, a UK-based company that develops AI applications for accident and disaster recovery.(55:26) Azin shared her excitement for ML applications using data-efficient methods to enhance life quality.(57:46) Closing segment.Azin’s Contact InfoWebsiteTwitterLinkedInGoogle ScholarGitHubMentioned ContentPublications“Barriers to Adoption of Information Technology in Healthcare” (2017)“Subspace Selection to Suppress Confounding Source Domain Information in AAM TransferLearning” (2017)“A Hybrid Instance-based Transfer Learning Method” (2018)“Prediction of Workplace Injuries” (2019)“Algorithmic Bias in Clinical Populations — Evaluating and Improving Facial Analysis Technology in Older Adults with Dementia” (2019)“Limitations and Biases in Facial Landmark Detection” (2019)Blog Posts“An Introduction to Transfer Learning” (Dec 2018)“Overcoming The Cold-Start Problem: How We Make Intractable Tasks Tractable” (April 2021)PeopleYoshua Bengio (Professor of Computer Science and Operations Research at University of Montreal)Geoffrey Hinton (Professor of Computer Science at University of Toronto)Louis-Philippe Morency (Associate Professor of Computer Science at Carnegie Mellon University)Book“Machine Learning: A Probabilistic Approach” (by Kevin Murphy)Note: Azin and her collaborator are going to give a talk at ODSC Europe 2021 in June about a Georgian’s project with a portfolio company, Tractable. They have written a short blog post about it too which you can find HERE.

    Episode 62: Leading Organizations Through Analytics Transformations with Gordon Wong

    Play Episode Listen Later Apr 28, 2021 75:09


    Show Notes(02:09) Gordon briefly talked about his undergraduate studying Psychology and Philosophy at Rutgers University in the early 90s.(03:24) Gordon reflected on the first decade of his career getting into database technologies.(05:34) Gordon discussed his predilection towards consulting, specifically his role in the professional services team at AB Initio Software in the early 2000s.(08:02) Gordon recalled the challenges of leading data warehousing initiatives at Smarter Travel Media and ClickSquared in the 2000s.(13:14) Gordon emphasized the advantage of a multi-tenant database over a traditional relational database.(18:30) Gordon recalled his one-year stint at Cervello, leading business intelligence implementations for their clients.(21:59) Gordon elaborated on his projects during his 3 years as the director of business intelligence infrastructure at Fitbit.(26:09) Gordon dived into his framework of choosing data tooling vendors while at Fitbit (and how he settled with a tiny startup called Snowflake back then).(30:02) Gordon provided recommendations for startups to be data-driven.(33:24) Gordon recalled practices to foster effective collaboration while managing the 3 teams of data engineering, data warehousing, and data analytics at Fitbit.(36:44) Gordon went over his proudest accomplishment as the director of data engineering at ezCater, making substantial improvements to their data warehouse platform.(38:59) Gordon shared his framework for interviewing data engineers.(41:39) Gordon walked through his consulting engagement in analytics engineering for Zipcar and data warehousing for edX.(46:17) Gordon reflected on his time as the Vice President of business intelligence at HubSpot.(50:50) Gordon unpacked his notion of “Data Hierarchy of Needs,” which entails the five pillars — data security, data quality, system reliability, user experience, and data coverage.(56:55) Gordon discussed current opportunities for driving better social outcomes and empowering democracy through data.(59:48) Gordon shared the key criteria that enable healthy team dynamics from his hands-on experience building data teams.(01:02:13) Gordon unpacked the central features and benefits of Snowflake for the un-initiated.(01:06:25) Gordon gave his verdict for the ETL tooling landscape in the next few years.(01:08:33) Gordon described the data community in Boston.(01:09:52) Closing segment.Gordon’s Contact InfoLinkedInMentioned ContentPeopleTristan Handy (co-founder of Fishtown Analytics and co-creator of dbt)Michael Kaminsky (who coined the term “Analytics Engineering”)Barr Moses (co-founder and CEO of Monte Carlo, who coined the term “Data Observability”)Book“Start With Why” (By Simon Sinek)

    Episode 61: Meta Reinforcement Learning with Louis Kirsch

    Play Episode Listen Later Apr 18, 2021 61:04


    Show Notes(2:05) Louis went over his childhood as a self-taught programmer and his early days in school as a freelance developer.(4:22) Louis described his overall undergraduate experience getting a Bachelor’s degree in IT Systems Engineering from Hasso Plattner Institute, a highly-ranked computer science university in Germany.(6:10) Louis dissected his Bachelor thesis at HPI called “Differentiable Convolutional Neural Network Architectures for Time Series Classification,” — which addresses the problem of automatically designing architectures for time series classification efficiently, using a regularization technique for ConvNet that enables joint training of network weights and architecture through back-propagation.(7:40) Louis provided a brief overview of his publication “Transfer Learning for Speech Recognition on a Budget,” — which explores Automatic Speech Recognition training by model adaptation under constrained GPU memory, throughput, and training data.(10:31) Louis described his one-year Master of Research degree in Computational Statistics and Machine Learning at the University College London supervised by David Barber.(12:13) Louis unpacked his paper “Modular Networks: Learning to Decompose Neural Computation,” published at NeurIPS 2018 — which proposes a training algorithm that flexibly chooses neural modules based on the processed data.(15:13) Louis briefly reviewed his technical report, “Scaling Neural Networks Through Sparsity,” which discusses near-term and long-term solutions to handle sparsity between neural layers.(18:30) Louis mentioned his report, “Characteristics of Machine Learning Research with Impact,” which explores questions such as how to measure research impact and what questions the machine learning community should focus on to maximize impact.(21:16) Louis explained his report, “Contemporary Challenges in Artificial Intelligence,” which covers lifelong learning, scalability, generalization, self-referential algorithms, and benchmarks.(23:16) Louis talked about his motivation to start a blog and discussed his two-part blog series on intelligence theories (part 1 on universal AI and part 2 on active inference).(27:46) Louis described his decision to pursue a Ph.D. at the Swiss AI Lab IDSIA in Lugano, Switzerland, where he has been working on Meta Reinforcement Learning agents with Jürgen Schmidhuber.(30:06) Louis created a very extensive map of reinforcement learning in 2019 that outlines the goal, methods, and challenges associated with the RL domain.(33:50) Louis unpacked his blog post reflecting on his experience at NeurIPS 2018 and providing updates on the AGI roadmap regarding topics such as scalability, continual learning, meta-learning, and benchmarks.(37:04) Louis dissected his ICLR 2020 paper “Improving Generalization in Meta Reinforcement Learning using Learned Objectives,” which introduces a novel algorithm called MetaGenRL, inspired by biological evolution.(44:03) Louis elaborated on his publication “Meta-Learning Backpropagation And Improving It,” which introduces the Variable Shared Meta-Learning framework that unifies existing meta-learning approaches and demonstrates that simple weight-sharing and sparsity in a network are sufficient to express powerful learning algorithms.(51:14) Louis expands on his idea to bootstrap AI that entails how the task, the general meta learner, and the unsupervised objective should interact (proposed at the end of his invited talk at NeurIPS 2020).(54:14) Louis shared his advice for individuals who want to make a dent in AI research.(56:05) Louis shared his three most useful productivity tips.(58:36) Closing segment.Louis’s Contact InfoWebsiteTwitterLinkedInGoogle ScholarGitHubMentioned ContentPapers and ReportsDifferentiable Convolutional Neural Network Architectures for Time Series Classification (2017)Transfer Learning for Speech Recognition on a Budget (2017)Modular Networks: Learning to Decompose Neural Computation (2018)Contemporary Challenges in Artificial Intelligence (2018)Characteristics of Machine Learning Research with Impact (2018)Scaling Neural Networks Through Sparsity (2018)Improving Generalization in Meta Reinforcement Learning using Learned Objectives (2019)Meta-Learning Backpropagation And Improving It (2020)Blog PostsTheories of Intelligence — Part 1 and Part 2 (July 2018)Modular Networks: Learning to Decompose Neural Computation (May 2018)How to Make Your ML Research More Impactful (Dec 2018)A Map of Reinforcement Learning (Jan 2019)NeurIPS 2018, Updates on the AI Roadmap (Jan 2019)MetaGenRL: Improving Generalization in Meta Reinforcement Learning (Oct 2019)General Meta-Learning and Variable Sharing (Nov 2020)PeopleJeff Clune (for his push on meta-learning research)Kenneth Stanley (for his deep thoughts on open-ended learning)Jürgen Schmidhuber (for being a visionary scientist)Book“Grit” (by Angela Duckworth)

    Episode 60: Algorithms and Data Structures for Massive Datasets with Dzejla Medjedovic

    Play Episode Listen Later Apr 5, 2021 71:25


    Show Notes(01:58) Dzejla described her undergraduate experience studying Computer Science at the Sarajevo School of Science and Technology back in the mid-2000s.(07:59) Dzejla recapped her overall experience getting a Ph.D. in Computer Science at Stony Brook University.(14:38) Dzejla unpacked the key research problem in her Ph.D. thesis titled “Upper and Lower Bounds on Sorting and Searching in External Memory.”(19:13) Dzejla went over the details of her paper “Don’t Thrash: How to Cache Your Hash on Flash,” — which describes the Cascade Filter, an approximate-membership-query data structure that scales beyond main memory, that is an alternative to the well-known Bloom-filter data structure.(24:41) Dzejla elaborated on her work “The batched predecessor problem in external memory,” — which studies the lower bounds in three external memory models: the I/O comparison model, the I/O pointer-machine model, and the index-ability model.(29:56) Dzejla shared her learnings from being a Teaching Assistant for the Introduction to Algorithms course at Stony Brook (both at the undergraduate and graduate level).(35:08) Dzejla went over her summer internships at Microsoft’s Server and Tools Division during her Ph.D.(41:06) Dzejla reasoned about her decision to return to Sarajevo School of Science and Technology as an Assistant Professor of Computer Science.(47:22) Dzejla dissected the essential concepts and methods covered in her Data Structures, Introductory Algorithms, Advanced Algorithms, and Algorithms for Big Data courses taught at SSIT.(48:42) Dzejla provided a brief overview of the Computer Science/Software Engineering department at the International University of Sarajevo (where she has been a professor since 2017.(50:57) Dzejla briefly talked about the courses that she taught at IUS, including Intro to Programming, Human-Computer Interaction, and Algorithms/Data Structures.(52:49) Dzejla shared the challenges of writing Algorithms and Data Structures for Massive Datasets, which introduces data processing and analytics techniques specifically designed for large distributed datasets.(56:14) Dzejla explained concepts in Part 1 of the book — including Hash Tables, Approximate Membership, Bloom Filters, Frequency/Cardinality Estimation, Count-Min Sketch, and Hyperloglog.(58:38) Dzejla provided a brief overview of techniques to handle streaming data in Part 2 of the book.(01:00:14) Dzejla mentioned the data structures for large databases and external-memory algorithms in Part 3 of the book.(01:02:15) Dzejla shared her thoughts about the tech community in Sarajevo.(01:04:16) Closing segment.Dzejla’s Contact InfoLinkedInTwitterGoogle ScholarMentioned ContentPapers“Upper and Lower Bounds on Sorting and Searching in External Memory” (Dzejla’s Ph.D. Thesis, 2014)“Don’t Thrash: How to Cache Your Hash on Flash” (2012)“The batched predecessor problem in external memory” (2014)PeopleErik Demaine (Computer Science Professor at MIT)Michael Bender (Computer Science Professor at Stony Brook, Dzejla’s Ph.D. Advisor)Joseph Mitchell (Computational Geometry Professor at Stony Brook)Steven Skiena (Computer Science Professor at Stony Brook)Jeff Erickson (Computer Science Professor at UIUC)Books“Algorithms and Data Structures for Massive Datasets” (by Dzejla Medjedovic, Emin Tahirovic, and Ines Dedovic)“The Algorithm Design Manual” (by Steven Skiena)Here is a permanent 40% discount code (good for all Manning products in all formats) for Datacast listeners: poddcast19. Link at http://mng.bz/4MAR.Here is one free eBook code good for a copy of Algorithms and Data Structures for Massive Datasets for a lucky listener: algdcsr-7135. Link at http://mng.bz/Q2y6

    Episode 59: Bridging The Gap Between Data and Models with Willem Pienaar

    Play Episode Listen Later Mar 24, 2021 48:57


    Show Notes(1:45) Willem discussed his undergraduate degree in Mechatronic Engineering at Stellenbosch University in the early 2010s.(2:34) Willem recalled his entrepreneurial journey founding and selling a networking startup that provides internet access to private residents on campus.(5:37) Willem worked for two years as a Software Engineer focusing on data systems at Systems Anywhere in Capetown after college.(6:49) Willem talked about his move to Bangkok working as a Senior Software Engineer at INDEFF, a company in industrial control systems.(9:52) Willem went over his decision to join Gojek, a leading Indonesian on-demand multi-service platform and digital payment technology group.(12:16) Willem mentioned the engineering challenges associated with building complex data systems for super-apps.(14:50) Willem dissected Gojek’s ML platform, including these four solutions for various stages of the ML life cycle: Clockwork, Merlin, Feast, and Turing.(19:24) Willem recapped the lessons from designing the ML platform to meet Gojek’s scaling requirements — as delivered at Cloud Next 2018.(23:09) Willem briefly went through the key design components to incorporate Kubeflow pipelines into Gojek’s existing ML platform — as delivered at KubeCon 2019.(26:21) Willem explained the inception of Feast, an open-source feature store that bridges the gap between data and models.(32:20) Willem talked about prioritizing the product roadmap and engaging the community for an open-source project.(35:07) Willem recapped the key lessons learned and envisioned Feast's future to be a lightweight modular feature store.(37:29) Willem explained the differences between commercial and open-source feature stores (given Tecton’s recent backing of Feast).(41:36) Willem reflected on his experience living and working in Southeast Asia.(44:33) Closing segment.Willem’s Contact InfoTwitterLinkedInGitHubMentioned ContentFeastFeast Project website: feast.devFeast Slack community: #FeastFeast Documentation: docs.feast.devFeast GitHub repository: feast-dev/feastFeast on StackOverflow: stackoverflow.com/questions/tagged/feastFeast Wiki: wiki.lfaidata.foundation/display/FEAST/Feast+HomeFeast Twitter: @feast_devArticleAn Introduction to Gojek’s Machine Learning Platform (2019)Introducing Feast: An Open-Source Feature Store For Machine Learning (2019)A State of Feast (2020)Why Tecton is Backing The Feast Open-Source Feature Store (2020)TalksLessons Learned Scaling Machine Learning at GoJek on Google Cloud (Cloud Next 2018)Accelerating Machine Learning App Development with Kubeflow Pipelines (Cloud Next 2019)Moving People and Products with Machine Learning on Kubeflow (KubeCon 2019)PeopleDavid Aronchick (Open-Source ML Strategy at Azure, Ex-PM for Kubernetes at Google, Co-Founder of Kubeflow, Advisor to Tecton)Jeremy Lewi (Principal Engineer at Primer.ai, Co-Founder of Kubeflow)Felipe Hoffa (Developer Advocate for BigQuery, Data Cloud Advocate for Snowflake)BookCal Newport’s “Deep Work”Willem will be a speaker at Tecton’s apply() virtual conference (April 21-22, 2021) for data and ML teams to discuss the practical data engineering challenges faced when building ML for the real world. Participants will share best practice development patterns, tools of choice, and emerging architectures they use to successfully build and manage production ML applications. Everything is on the table from managing labeling pipelines, to transforming features in real-time, and serving at scale. Register for free now: https://www.applyconf.com/!

    Episode 58: Deep Learning Meets Distributed Systems with Jim Dowling

    Play Episode Listen Later Mar 19, 2021 79:15


    Show Notes(1:56) Jim went over his education at Trinity College Dublin in the late 90s/early 2000s, where he got early exposure to academic research in distributed systems.(4:26) Jim discussed his research focused on dynamic software architecture, particularly the K-Component model that enables individual components to adapt to a changing environment.(5:37) Jim explained his research on collaborative reinforcement learning that enables groups of reinforcement learning agents to solve online optimization problems in dynamic systems.(9:03) Jim recalled his time as a Senior Consultant for MySQL.(9:52) Jim shared the initiatives at the RISE Research Institute of Sweden, in which he has been a researcher since 2007.(13:16) Jim dissected his peer-to-peer systems research at RISE, including theoretical results for search algorithm and walk topology.(15:30) Jim went over challenges building peer-to-peer live streaming systems at RISE, such as GradientTV and Glive.(18:18) Jim provided an overview of research activities at the Division of Software and Computer Systems at the School of Electrical Engineering and Computer Science at KTH Royal Institute of Technology.(19:04) Jim has taught courses on Distributed Systems and Deep Learning on Big Data at KTH Royal Institute of Technology.(22:20) Jim unpacked his O’Reilly article in 2017 called “Distributed TensorFlow,” which includes the deep learning hierarchy of scale.(29:47) Jim discussed the development of HopsFS, a next-generation distribution of the Hadoop Distributed File System (HDFS) that replaces its single-node in-memory metadata service with a distributed metadata service built on a NewSQL database.(34:17) Jim rationalized the intention to commercialize HopsFS and built Hopsworks, an user-friendly data science platform for Hops.(36:56) Jim explored the relative benefits of public research money and VC-funded money.(41:48) Jim unpacked the key ideas in his post “Feature Store: The Missing Data Layer in ML Pipelines.”(47:31) Jim dissected the critical design that enables the Hopsworks feature store to refactor a monolithic end-to-end ML pipeline into separate feature engineering and model training pipelines.(52:49) Jim explained why data warehouses are insufficient for machine learning pipelines and why a feature store is needed instead.(57:59) Jim discussed prioritizing the product roadmap for the Hopswork platform.(01:00:25) Jim hinted at what’s on the 2021 roadmap for Hopswork.(01:03:22) Jim recalled the challenges of getting early customers for Hopsworks.(01:04:30) Jim intuited the differences and similarities between being a professor and being a founder.(01:07:00) Jim discussed worrying trends in the European Tech ecosystem and the role that Logical Clocks will play in the long run.(01:13:37) Closing segment.Jim’s Contact InfoLogical ClocksTwitterLinkedInGoogle ScholarMediumACM ProfileGitHubMentioned ContentResearch Papers“The K-Component Architecture Meta-Model for Self-Adaptive Software” (2001)“Dynamic Software Evolution and The K-Component Model” (2001)“Using feedback in collaborative reinforcement learning to adaptively optimize MANET routing” (2005)“Building Autonomic Systems Using Collaborative Reinforcement Learning” (2006)“Improving ICE Service Selection in a P2P System using the Gradient Topology” (2007)“gradienTv: Market-Based P2P Live Media Streaming on the Gradient Overlay” (2010)“GLive: The Gradient Overlay as a Market Maker for Mesh-Based P2P Live Streaming” (2011)“HopsFS: Scaling Hierarchical File System Metadata Using NewSQL Databases” (2016)“Scaling HDFS to More Than 1 Million Operations Per Second with HopsFS” (2017)“Hopsworks: Improving User Experience and Development on Hadoop with Scalable, Strongly Consistent Metadata” (2017)“Implicit Provenance for Machine Learning Artifacts” (2020)“Time Travel and Provenance for Machine Learning Pipelines” (2020)“Maggy: Scalable Asynchronous Parallel Hyperparameter Search” (2020)Articles“Distributed TensorFlow” (2017)“Reflections on AWS’s S3 Architectural Flaws” (2017)“Meet Michelangelo: Uber’s Machine Learning Platform” (2017)“Feature Store: The Missing Data Layer in ML Pipelines” (2018)“What Is Wrong With European Tech Companies?” (2019)“ROI of Feature Stores” (2020)“MLOps With A Feature Store” (2020)“ML Engineer Guide: Feature Store vs. Data Warehouse” (2020)“Unifying Single-Host and Distributed Machine Learning with Maggy” (2020)“How We Secure Your Data With Hopsworks” (2020)“One Function Is All You Need For ML Experiments” (2020)“Hopsworks: World’s Only Cloud-Native Feature Store, now available on AWS and Azure” (2020)“Hopsworks 2.0: The Next Generation Platform for Data-Intensive AI with a Feature Store” (2020)“Hopsworks Feature Store API 2.0, a new paradigm” (2020)“Swedish startup Logical Clocks takes a crack at scaling MySQL backend for live recommendations” (2021)ProjectsApache Hudi (by Uber)Delta Lake (by Databricks)Apache Iceberg (by Netflix)MLflow (by Databricks)Apache Flink (by The Apache Foundation)PeopleLeslie Lamport (The Father of Distributed Computing)Jeff Dean (Creator of MapReduce and TensorFlow, Lead of Google AI)Richard Sutton (The Father of Reinforcement Learning — who wrote “The Bitter Lesson”)Programming BooksC++ Programming Languages books (by Scott Meyers)“Effective Java” (by Joshua Bloch)“Programming Erlang” (by Joe Armstrong)“Concepts, Techniques, and Models of Computer Programming” (by Peter Van Roy and Seif Haridi)

    Episode 57: Building Data Science Projects with Pier Paolo-Ippolito

    Play Episode Listen Later Mar 6, 2021 54:59


    Show Notes(2:20) Pier shared his college experience at the University of Southampton studying Electronic Engineering.(3:46) For his final undergraduate project, Pier developed a suite of games and used machine learning to analyze brainwaves data that can classify whether a child is affected or not by autism.(11:26) Pier went over his favorite courses and involvement with the AI Society during his additional year at the University of Southampton to get a Master’s in Artificial Intelligence.(13:40) For his Master’s thesis called “Causal Reasoning in Machine Learning,” Pier created and deployed a suite of Agent-Based and Compartmental Models to simulate epidemic disease developments in different types of communities.(26:51) Pier went over his stints as a developer intern at Fidessa and a freelance data scientist at Digital-Dandelion.(29:21) Pier reflected on his time (so far) as a data scientist at SAS Institute, where he helps their customers solve various data-driven challenges using cloud-based technologies and DevOps processes.(33:37) Pier discussed the key benefits that writing and editing technical content for Towards Data Science to his professional development.(36:31) Pier covered the threads that he kept pulling with his blog posts.(38:50) Pier talked about his Augmented Reality Personal Business Card created in HTML using the AR.js library.(41:12) Pier brought up data structures in two other impressive JavaScript projects using TensorFlow.js and ml5.js.(44:19) Pier went over his experience working with data visualization tools such as Plotly, R Shiny, and Streamlit.(47:27) Pier talked about his work on a chapter for a book called “Applied Data Science in Tourism” that is going to be published with Springer this year.(48:37) Pier shared his thoughts regarding the tech community in London.(49:19) Closing segment.Pier’s Contact InfoWebsiteLinkedInTwitterGitHubMediumPatreonKaggleMentioned Content“Alleviate Children’s Health Issues Through Games and Machine Learning”“Causal Reasoning in Machine Learning”Andrej Karpathy (Director of AI and Autopilot at Tesla)Cassie Kozyrkov (Chief Decision Scientist at Google)Iain Brown (Head of Data Science at SAS)“The Book Of Why” (By Judea Pearl)“Pattern Recognition and Machine Learning” (by Christopher Bishop)

    Episode 56: Apprehending Quantum Computation with Alba Cervera-Lierta

    Play Episode Listen Later Feb 21, 2021 77:26


    Timestamps(1:55) Alba shared her background growing up interested in studying Physics and pivoting into quantum mechanics.(3:33) Alba went over her Bachelor’s in Fundamental Physics at The University of Barcelona.(4:54) Alba continued her education with an M.S. degree that specialized in Particle Physics and Gravitation.(6:40) Alba started her Ph.D. in Physics in 2015 and discussed her first publication, “Operational Approach to Bell Inequalities: Application to Qutrits.”(9:48) Alba also spent time as a visiting scholar at the University of Oxford and the University of Madrid during her Ph.D.(11:25) Alba explained her second paper to understand the connection between maximal entanglement and the fundamental symmetries of high-energy physics.(13:27) Alba dissected her next work titled “Multipartite Entanglement in Spin Chains and The Hyperdeterminant.”(18:56) Alba shared the origin of Quantic, a quantum computation joint effort between the University of Barcelona and the Barcelona Supercomputing Center.(22:27) Alba unpacked her article “Quantum Computation: Playing The Quantum Symphony,” making a metaphor between quantum computing and musical symphony.(27:47) Alba discussed the motivation and contribution of her paper “Exact Ising Model Simulation On A Quantum Computer.”(32:51) Alba recalled creating a tutorial that ended up winning the Teach Me QISKit challenge from IBM back in 2018.(35:01) Alba elaborated on her paper “Quantum Circuits For the Maximally Entangled States,” which designs a series of quantum circuits that generate absolute maximally entangled states to benchmark a quantum computer.(38:54) Alba dissected key ideas in her paper “Data Re-Uploading For a Universal Quantum Classifier.”(43:51) Alba explained how she leveled up her knowledge of classical neural networks.(47:40) Alba shared her experience as a Postdoctoral Fellow at The Matter Lab at the University of Toronto — working on quantum machine learning and variational quantum algorithms (checked out the Quantum Research Seminars Toronto that she has been organizing).(52:18) Alba explained her work on the Meta-Variational Quantum Eigensolver algorithm capable of learning the ground state energy profile of a parametrized Hamiltonian.(59:23) Alba went over Tequila, a development package for quantum algorithms in Python that her group created.(01:04:49) Alba presented a quantum calling for new algorithms, applications, architectures, quantum-classical interface, and more (as presented here).(01:08:59) Alba has been active in education and public outreach activities about encouraging scientific vocations for young minds, especially in Catalonia.(01:12:07) Closing segment.Her Contact InfoWebsiteTwitterLinkedInGoogle ScholarGitHubHer Recommended ResourcesEwin Tang (Ph.D. Student in Theoretical Computer Science at the University of Washington)Alán Aspuru-Guzik (Professor of Chemistry and Computer Science at the University of Toronto, Alba’s current supervisor)José Ignacio Latorre (Professor of Theoretical Physics at the University of Barcelona, Alba’s former supervisor)Quantum Computation and Quantum Information (by Michael Nielsen and Isaac Chuang)Quantum Field Theory and The Standard Model (by Matthew Schwarz)The Structure of Scientific Revolutions (by Thomas Kuhn)Against Method (by Paul Feyerabend)Quantum Computing Since Democritus (by Scott Aaronson)

    Episode 55: Making Apache Spark Developer-Friendly and Cost-Effective with Jean-Yves Stephan

    Play Episode Listen Later Feb 11, 2021 52:02


    Timestamps(2:07) JY discussed his college time studying Computer Science and Applied Math at Ecole Polytechnique — a leading French institute in science and technology.(3:04) JY reflected on time at Stanford getting a Master’s in Management Science and Engineering, where he served as a Teaching Assistant for CS 229 (Machine Learning) and CS 246 (Mining Massive Datasets).(6:14) JY walked over his ML engineering internship at LiveRamp — a data connectivity platform for the safe and effective use of data.(7:54) JY reflected on his next three years at Databricks, first as a software engineer and then as a tech lead for the Spark Infrastructure team.(10:00) JY unpacked the challenges of packaging/managing/monitoring Spark clusters and automating the launch of hundreds of thousands of nodes in the cloud every day.(14:48) JY shared the founding story behind Data Mechanics, whose mission is to give superpowers to the world's data engineers so they can make sense of their data and build applications at scale on top of it.(18:09) JY explained the three tenets of Data Mechanics: (1) managed and serverless, (2) integrated into clients’ workflows, and (3) built on top of open-source software (read the launch blog post).(22:06) JY unpacked the core concepts of Spark-On-Kubernetes and evaluated the benefits/drawbacks of this new deployment mode — as presented in “Pros and Cons of Running Apache Spark on Kubernetes.”(26:00) JY discussed Data Mechanics’ main improvements on the open-source version of Spark-On-Kubernetes — including an intuitive user interface, dynamic optimizations, integrations, and security — as explained in “Spark on Kubernetes Made Easy.”(28:35) JY went over Data Mechanics Delight, a customized Spark UI which was recently open-sourced.(35:40) JY shared the key ideas in his thought-leading piece on how to be successful with Apache Spark in 2021.(38:42) JY went over his experience going through the Y Combinator program in summer 2019.(40:56) JY reflected on the key decisions to get the first cohort of customers for Data Mechanics.(42:26) JY shared valuable hiring lessons for early-stage startup founders.(44:34) JY described the data and tech community in France.(47:19) Closing segment.His Contact InfoTwitterLinkedInData MechanicsHis Recommended ResourcesJure Leskovec (Associate Professor of Computer Science at Stanford / Chief Scientist at Pinterest)Jeff Bezos (Founder of Amazon)Matei Zaharia (CTO of Databricks and creator of Apache Spark)“Designing For Data-Intensive Applications” (by Martin Kleppmann)

    Episode 54: Information Retrieval Research, Data Science For Space Missions, and Open-Source Software with Chris Mattmann

    Play Episode Listen Later Feb 4, 2021 82:43


    Timestamps(2:55) Chris went over his experience studying Computer Science at the University of Southern California for undergraduate in the late 90s.(5:26) Chris recalled working as a Software Engineer at NASA Jet Propulsion Lab in his sophomore year at USC.(9:54) Chris continued his education at USC with an M.S. and then a Ph.D. in Computer Science. Under the guidance of Dr. Nenad Medvidović, his Ph.D. thesis is called “Software Connectors For Highly-Distributed And Voluminous Data-Intensive Systems.” He proposed DISCO, a software architecture-based systematic framework for selecting software connectors based on eight key dimensions of data distribution.(16:28) Towards the end of his Ph.D., Chris started getting involved with the Apache Software Foundation. More specifically, he developed the original proposal and plan for Apache Tika (a content detection and analysis toolkit) in collaboration with Jérôme Charron to extract data in the Panama Papers, exposing how wealthy individuals exploited offshore tax regimes.(24:58) Chris discussed his process of writing “Tika In Action,” which he co-authored with Jukka Zitting in 2011.(27:01) Since 2007, Chris has been a professor in the Department of Computer Science at USC Viterbi School of Engineering. He went over the principles covered in his course titled “Software Architectures.”(29:49) Chris touched on the core concepts and practical exercises that students could gain from his course “Information Retrieval and Web Search Engines.”(32:10) Chris continued with his advanced course called “Content Detection and Analysis for Big Data” in recent years (check out this USC article).(36:31) Chris also served as the Director of the USC’s Information Retrieval and Data Science group, whose mission is to research and develop new methodology and open source software to analyze, ingest, process, and manage Big Data and turn it into information.(41:07) Chris unpacked the evolution of his career at NASA JPL: Member of Technical Staff -> Senior Software Architect -> Principal Data Scientist -> Deputy Chief Technology and Innovation Officer -> Division Manager for the AI, Analytics, and Innovation team.(44:32) Chris dove deep into MEMEX — a JPL’s project that aims to develop software that advances online search capabilities to the deep web, the dark web, and nontraditional content.(48:03) Chris briefly touched on XDATA — a JPL’s research effort to develop new computational techniques and open-source software tools to process and analyze big data.(52:23) Chris described his work on the Object-Oriented Data Technology platform, an open-source data management system originally developed by NASA JPL and then donated to the Apache Software Foundation.(55:22) Chris shared the scientific challenges and engineering requirements associated with developing the next generation of reusable science data processing systems for NASA’s Orbiting Carbon Observatory space mission and the Soil Moisture Active Passive earth science mission.(01:01:05) Chris talked about his work on NASA’s Machine Learning-based Analytics for Autonomous Rover Systems — which consists of two novel capabilities for future Mars rovers (Drive-By Science and Energy-Optimal Autonomous Navigation).(01:04:24) Chris quantified the Apache Software Foundation's impact on the software industry in the past decade and discussed trends in open-source software development.(01:07:15) Chris unpacked his 2013 Nature article called “A vision for data science” — in which he argued that four advancements are necessary to get the best out of big data: algorithm integration, development and stewardship, diverse data formats, and people power.(01:11:54) Chris revealed the challenges of writing the second edition of “Machine Learning with TensorFlow,” a technical book with Manning that teaches the foundational concepts of machine learning and the TensorFlow library's usage to build powerful models rapidly.(01:15:04) Chris mentioned the differences between working in academia and industry.(01:16:20) Chris described the tech and data community in the greater Los Angeles area.(01:18:30) Closing segment.His Contact InfoWikipediaNASA PageGoogle ScholarUSC PageTwitterLinkedInGitHubHis Recommended ResourcesDoug Cutting (Founder of Lucene and Hadoop)Hilary Mason (Ex Data Scientist at bit.ly and Cloudera)Jukka Zitting (Staff Software Engineer at Google)"The One Minute Manager" (by Ken Blanchard and Spencer Johnson)

    Episode 53: Algorithms and Data Structures In Action with Marcello La Rocca

    Play Episode Listen Later Jan 25, 2021 62:49


    Show Notes(2:09) Marcello described his academic experience getting a Master’s Degree in Computer Science from the Universita di Catania in the early 2000s, where his thesis is called Evolutionary Randomized Graph Embedder.(6:14) Marcello commented on his career phase working as a web developer across various places in Europe.(9:18) Marcello discussed his time working as a software engineer at INPS, a government-owned company that now handles most Italian citizens' pubic-related data.(10:42) Marcello talked about his time as a data visualization engineer at SwiftIQ. He created a data visualization library that allows the inclusion of dynamic charts in HTML pages with just a few JavaScript lines.(13:40) Marcello went over his projects while working as a full-stack software engineer for Twitter’s User Services Engineering team in Dublin.(17:19) Marcello reflected on his time at Microsoft Zurich’s Social and Engagement team, contributing to machine learning infrastructure and tools.(21:28) Marcello briefly touched on his one-year stint at Apple Zurich as a Senior Applied Research Engineer.(23:49) Marcello talked about the challenges while writing “Algorithms and Data Structures in Action,” which introduces a diverse range of algorithms used in web apps, systems programming, and data manipulation.(27:11) Marcello expanded upon part 1 of the book, including advanced data structures such as D-ary Heaps, Randomized Treaps, Bloom Filters, Disjoint Sets, Tries/Radix Trees, and Cache.(34:51) Marcello brought up data structures to perform efficient multi-dimensional queries, including various nearest neighbor searches and clustering techniques, in part 2 of the book.(39:21) Marcello briefly described the algorithms in part 3 of the book — graph embeddings, gradient descent, simulated annealing, and genetic algorithms.(48:28) Marcello talked about his work on jsgraph — a lightweight library to model graphs, run graphs algorithms, and display them on screen.(52:06) Marcello compared Python, Java, and JavaScript programming languages.(54:13) Marcello discussed his current interest in quantum computing.(56:18) Marcello shared his thoughts regarding Dublin, Zurich, and Rome's tech communities.(57:37) Closing segment.His Contact InfoTwitterLinkedInGitHubBlogHis Recommended Resources"Algorithms and Data Structures in Action" (Marcello's book with Manning)Andrew NgGeoffrey HintonFrancois Chollet"Scalability Rules" (by Martin Abbott and Michael Fischer)This is the 40% discount code that is good for all Manning’s products in all formats: poddcast19.These are 5 free eBook codes, each good for one copy of “Algorithms and Data Structures in Action”:adsdcr-5E76adsdcr-EE51adsdcr-DD47adsdcr-B1BFadsdcr-A61F

    Episode 52: Graph Databases In Action with Dave Bechberger

    Play Episode Listen Later Jan 17, 2021 61:58


    Show Notes(2:10) Dave talked briefly about his Electrical Engineering study at Rensselaer Polytechnic Institute back in the late 90s.(4:03) Dave commented on his career phase working as a software engineer across various companies in Bozeman, Montana.(7:38) Dave discussed his work as a senior architect and tech lead at Expero, a Houston-based startup that develops custom software exclusively for domain-expert users.(11:26) Dave briefly defined common big data frameworks (Hadoop, Apache Spark) and databases (Apache Cassandra, Apache Kafka).(13:37) Dave went over the challenges during his time as a chief software architect at Gene by Gene, a biotech company focusing on DNA-based ancestry and genealogy.(20:00) Dave shared the common patterns and anti-patterns of using graph databases (in reference to his talk “A Practical Guide to Graph Databases”).(26:16) Dave walked through the three categories of graph technologies: Graph Computing Engine, RDF TripleStore, and Labeled Property Graph (in reference to his talk “A Skeptics Guide to Graph Databases”).(33:03) Dave discussed his move to DataStax’s Global Graph Practice team as a solutions architect and graph database subject matter expert.(36:00) Dave explained the design of DataStax’s enterprise solution called Customer 360, which collapses data silos to drive business value.(41:16) Dave talked about his current experience as a Senior Graph Architect at AWS.(43:51) Dave mentioned the challenges while writing "Graph Databases In Action" (published last October).(47:25) Dave explained the open-source Apache TinkerPop framework and the Gremlin language used in the book for the uninitiated.(51:04) Dave discussed trends in big data and distributed systems that he is most excited about.(55:06) Closing segment.His Contact InfoWebsiteTwitterLinkedInGitHubHis Recommended Resources"Graph Databases In Action" (Associated Code Repository)Martin Fowler (Founder of ThoughtWorks)Martin Kleppmann (Author of "Designing Data-Intensive Applications")Andrew Ng (Professor at Stanford, Co-Founder of Google Brain and Coursera, Ex-Chief Scientist at Baidu)"Pragmatic Programmer" (by Andy Hunt and Dave Thomas)"The Five Dysfunctions Of A Team" (by Patrick Lencioni)"How To Observe Scientific Advice for Common Real-World Problems" (by Randall Munroe)This is the 40% discount code that is good for all Manning's products in all formats: poddcast19.These are 5 free eBook codes, each good for one copy of "Graph Databases In Action":gdadcr-E55Fgdadcr-B896gdadcr-8C53gdadcr-AAE1gdadcr-39F0

    Episode 51: Research and Tooling for Computer Vision Systems with Jason Corso

    Play Episode Listen Later Jan 8, 2021 80:33


    Show Notes(2:13) Jason went over his experience studying Computer Science at Loyola College in Baltimore for undergraduate, where he got an early exposure to academic research in image registration.(4:31) Jason described his graduate school experience at John Hopkins University, where he completed his Ph.D. on “Techniques for Vision-Based Human-Computer Interaction” that proposed the Visual Interaction Cues paradigm.(9:31) During his time as a Post-Doc Fellow at UCLA, Jason helped develop automatic segmentation and recognition techniques for brain tumors to improve the accuracy of diagnosis and treatment accuracy(14:27) From 2007 to 2014, Jason was a professor in the Computer Science and Engineering department at SUNY-Buffalo. He covered the content of two graduate-level courses on Bayesian Vision and Intro to Pattern Recognition that he taught.(18:20) On the topic of metric learning, Jason proposed an approach to data analysis and modeling for computer vision called "Active Clustering."(21:35) On the topic of image understanding, Jason created Generalized Image Understanding - a project that examined a unified methodology that integrates low-, mid-, and high-level elements for visual inference (equivalent to image captioning today).(24:51) On the topic of video understanding, Jason worked on ISTARE: Intelligent Spatio-Temporal Activity Reasoning Engine, whose objective is to represent, learn, recognize, and reason over activities in persistent surveillance videos.(27:46) Jason dissected Action Bank - a high-level representation of activity in video, which comprises of many individual action detectors sampled broadly in semantic space and viewpoint space.(35:30) Jason unpacked LIBSVX - a library of super voxel and video segmentation methods coupled with a principled evaluation benchmark based on quantitative 3D criteria for good super voxels.(40:06) Jason gave an overview of AI research activities at the University of Michigan, where he was a professor of Electrical Engineering and Computer Science from 2014 to 2020.(41:09) Jason covered the problems and projects in his graduate-level courses on Foundations of Computer Vision and Advanced Topics in Computer Vision at Michigan.(44:56) Jason went over his recent research on video captioning and video description.(47:03) Jason described his exciting software called BubbleNets, which chooses the best video frame for a human to annotate.(51:44) Jason shared anecdotes of Voxel51's inception and key takeaways that he has learned.(01:05:25) Jason talked about Voxel51's Physical Distancing Index that tracks the coronavirus global pandemic's impact on social behavior.(01:07:47) Jason discussed his exciting new chapter as the new director of the Stevens Institute for Artificial Intelligence.(01:11:28) Jason identified the differences and similarities between being a professor and being a founder.(01:14:55) Jason gave his advice to individuals who want to make a dent in AI research.(01:16:14) Jason mentioned the trends in computer vision research that he is most excited about at the moment.(01:17:23) Closing segment.His Contact InfoWikipediaGoogle ScholarWebsiteTwitterLinkedInHis Recommended ResourcesBubblenets: Video Object Segmentation for Computer VisionVoxel51's FiftyOne Open-Sourced LibraryJeff Siskind (Professor at Purdue University)CJ Taylor (Professor at the University of Pennsylvania)Kristen Grauman (Professor at the University of Austin)"An Introduction to Mathematical Statistics"

    Episode 50: Reducing Data Downtime with Barr Moses

    Play Episode Listen Later Dec 25, 2020 54:10


    Show Notes(2:23) Barr discussed growing up in Israel and serving as a commander of the Data Analytics unit at the Israeli Air Force.(4:10) Barr reflected on her college experience at Stanford studying Math and Computational Science.(7:24) Barr walked over the two career lessons learned from being a Management Consultant at Bain and Company.(9:51) Barr reflected on her time as VP of Customer Operations at Gainsight, which offers enterprise solutions for Customer Success and Product teams. She helped build and scale a global team covering various functions such as business operations, customer success, professional services.(12:32) Barr unpacked the notion of data downtime, introduced in her blog post “The Rise of Data Downtime.”(17:25) Barr unveiled the four main steps in the data reliability maturity curve: reactive, proactive, automated, and scalable - as indicated in “Closing The Data Downtime Gap."(21:09) Barr shared the founding story behind Monte Carlo, whose mission is to accelerate the world’s adoption of data by reducing data downtime.(24:29) Barr explained the five pillars of data observability.(27:45) Barr unpacked the rise of data catalogs as a powerful tool for data governance, along with the three categories of data catalog solutions that data teams are adopting - as presented in “What We Got Wrong About Data Governance.”(31:32) Barr discussed the benefits of using Data Mesh - a type of data platform architecture that embraces data ubiquity in the enterprise by leveraging a domain-oriented, self-serve design.(37:28) Barr went over a framework that looks at the business functions and the nature of the work to score the impact and allocate the ROI of the data team - as proposed in "Measuring the ROI of Your Data Organization."(40:39) Barr shared five practices for designing a platform that maximizes data's value and impact inside an organization.(43:27) Barr reflected on the key decisions to get the first cohort of customers for Monte Carlo.(46:31) Barr shared valuable hiring lessons.(48:48) Barr went over helpful resources throughout her journey as a founder.(50:13) Barr dropped the final advice for founders on seeking the right investors.(51:18) Closing segment.Her Contact InfoTwitterLinkedInMediumMonte CarloHer Recommended ResourcesStanford's "Mathematics and Magic Tricks" course (taught by Persi Diaconis)"The Biggest Bluff" by Maria KonnikovaSnowflakeDJ Patil (Former U.S. Chief Data Scientist)Monte Carlo's Customers

    Episode 49: Computational Neuroscience, Quantitative Finance, and Churn Prediction with Carl Gold

    Play Episode Listen Later Dec 14, 2020 68:37


    Show Notes(1:57) Carl recalled his undergraduate experience studying Electrical Engineering at Stanford back in the early 90s.(3:58) Carl recalled his graduate experience pursuing Master’s degrees in Computer Science at NYU and King’s College in the late 90s. For his Master's Thesis, he investigated Support Vector Machines with a Bayesian algorithm programmed in C.(6:45) Carl walked over his Ph.D. work in Computation and Neural Systems at CalTech, where he did a thesis on Biophysics of Extracellular Action Potentials.(13:11) Carl provided brief thoughts about his experience working as a business analyst and consultant for HBO during his Ph.D. period.(14:55) Carl went over his rationale behind his decision to move from academic neuroscience to quantitative finance.(19:19) Carl discussed his proudest accomplishments and valuable lessons learned from spending seven years at Morgan Stanley Capital International and rising to a leadership role as Vice President of Risk Modeling.(23:17) Carl uncovered his move to San Francisco to work as a lead data scientist at Sparked back in 2014, which builds a customer success SaaS solution.(27:10) Adding to his move to Zuora in 2015, Carl explained how the subscription business model works in layman terms.(31:44) Carl unpacked the common patterns that he saw from analyzing subscriber churn for companies across industries due to his work on Zuora Analytics.(33:30) Carl shared the process of creating the Subscription Economy Index, Zuora’s landmark index tracking the rapid ascent of the Subscription Economy, and distilled the key trends of the 2020 edition.(39:59) Carl unpacked the three reasons that make churn hard to fight: (1) Churn is hard to predict, (2) Churn is harder to prevent, and (3) Churn requires a multi-team effort (Watch his talks at the 2019 Data Council San Francisco and the 2020 Subscribed Online Conference).(44:46) Carl shared advice for data scientists who want to collaborate more effectively with other functional departments.(46:30) Carl emphasized the importance of creating great customer metrics, which are ratios of basic behavioral metrics to fight churn effectively.(53:49) Carl went over the challenges of writing “Fighting Churn With Data,” which provides a clear overview of churn concepts, along with hands-on tricks and tips developed through years of experience analyzing customer behavior.(55:53) Carl reflected on how his academic background in computational neuroscience contributes to his success as a quant analyst and a data scientist.(59:37) Carl compared his experience living and working across Los Angeles, New York, and San Francisco.(01:02:04) Closing segment.His Contact InfoLinkedInTwitterGitHubGoogle ScholarMediumHis Recommended Resources“Fighting Churn With Data” by Carl GoldKonrad Kording (Professor of Computational Neuroscience at the University of Pennsylvania)Kate Crawford (Distinguished Research Professor in Tech, Culture, and Society at New York University)Cassie Kozyrkov (Chief Decision Scientist at Google)"Freakonomics" by Stephen Dubner and Stephen LevittCarl's other podcast appearancesHere are the discount codes that you can use to purchase "Fighting Churn with Data" with 40% off:fcddcr-6D84fcddcr-4AE7fcddcr-6D9Cfcddcr-30BFfcddcr-9705

    Episode 48: AI Ethics, Open Data, and Recommendations Fairness with Jessie Smith

    Play Episode Listen Later Nov 27, 2020 57:46


    Show Notes(2:08) Jess discussed her foray into studying Software Engineering at California Polytechnic State University during college and revealed her favorite course on Computer Science Ethics taken there.(4:31) Jess unpacked her argument that it is important to shift the engineering mindset away from only asking how  to ask why - referring to his blog post “Changing The Engineer’s Mindset.”(7:27) Jess went over her summer internship experience at GoDaddy as a software engineer.(11:39) Jess talked about her time working as a research assistant for the Ethics and Emerging Sciences Group at Cal Poly, where she examined the ethical implications of AI “predictive policing” systems and survey the current role of fairness metrics for battling algorithmic bias.(16:27) Jess revealed her experience being involved with the open data movement in Colombia (read her articles “The Truth About Open Data” and “How To Use Data Science For Social Impact”).(24:22) Jess emphasized the importance of education to spread data literacy in developing nations.(26:35) Jess discussed her experience as a current Ph.D. student in the Department of Information Science at the University of Colorado, Boulder, where you focus on value tradeoffs in technology and machine learning ethics.(32:01) Jess unpacked the ETHItechniCAL framework to assist with ethical decision-making that she proposes in “The Trolley Problem Isn’t Theoretical Anymore.”(35:39) Jess unpacked her argument, saying that computer scientists must be educated to code with social responsibility and equipped with the correct tools to do so - as indicated in “How Tech Shapes Society.”(39:00) Jess discussed the work “Investigating Potential Factors Associated with Gender Discrimination in Collaborative Recommender Systems” with Masoud Mansoury and Himan Abdollahpouri.(42:54) Jess discussed the work “Exploring User Opinions of Fairness in Recommender Systems” with Nasim Sonboli.(47:12) Via her podcast The Radical AI, Jess unpacked the underrated AI and social issues that she came across.(49:17) Via her YouTube show Sci-Fi in Real Life, Jess shared her 3 favorite videos: "Dying To Be Alive," "Living On The Edge," and "Black Mirror Meta Episode."(52:25) Jess dug deep into her mission of cultivating positive social impacts for the world.(54:32) Closing segment.Her Contact InfoWebsiteTwitterMediumLinkedInGitHubRadical AI PodcastSci-Fi In Real Life YouTube ShowHer Recommended ResourcesUC Boulder's Internet Rules LabUC Boulder's That Recommender Systems LabSafiya NobleCathy O'NeilRuha Benjamin"The Courage To Be Disliked" by Ichiro Kishimi and Fumitake Koga

    Episode 47: Math and Machine Learning In Pedestrian Terms with Luis Serrano

    Play Episode Listen Later Nov 9, 2020 51:25


     Show Notes(2:12) Luis shared how he got excited about learning mathematics and specialized in combinatorics.(4:26) Luis discussed his experience studying Math for his Bachelor’s and Master’s degrees at the University of Waterloo  - where he took many courses in combinatorics and engaged in undergraduate research.(5:59) Luis pursued his Ph.D. in Mathematics at the University of Michigan - where he worked on Schubert Calculus that intersects combinatorics and geometry (check out his Ph.D. dissertation).(8:45) Luis distinguished the differences between doing research in mathematics and machine learning.(11:33) Luis went over his time as a Postdoc Fellow and Lecturer at the University of Quebec at Montreal - where he was a member of the LaCIM lab (whose areas of research originating in Combinatorics and its relationships to Algebra and Computer Science) and taught classes in French.(13:47) Luis explained why he left academia and got his job as a Machine Learning Engineer at Google in 2014.(16:33) Luis discussed the engineering and analytical challenges he encountered as part of the video recommendations team at YouTube.(19:58) Luis shared lessons he learned to transition from academia to industry.(22:25) Luis went over his move to become the Head of Content for AI and Data Science at Udacity, alongside his online education passion.(26:08) Luis explained Udacity's educational approach to course content design in various nano degree programs, including Machine Learning, Deep Learning, and Data Science.(28:46) Luis unpacked his end-to-end process of making YouTube, where he teaches concepts in Machine Learning and Math in layman terms.(31:01) Luis unpacked his statement, "Humans are bad at abstraction, but great at math," from his video “You Are Much Better At Math Than You Think.”(34:46) Luis shared his 3 favorite Machine Learning videos: Restricted Boltzmann Machines, A Friendly Introduction to Machine Learning, and My Story with the Thue-Morse Sequence.(37:18) Luis discussed the data science culture at Apple, where he spent one-year teaching machine learning to the employees and doing internal consulting in AI-related projects.(39:06) Luis revealed his interest in quantum computing. He works as a Quantum AI Research Scientist at Zapata Computing, a quantum software company that offers computing solutions for industrial and commercial use.(43:19) Luis mentioned the challenges of writing “Grokking Machine Learning” - a technical book with Manning planned to be published next year - like a mystery novel.(46:12) Luis shared the differences between working in Silicon Valley and Canada.(47:50) Closing segment.His Contact InfoWebsiteTwitterLinkedInYouTubeGitHubGoogle ScholarMediumHis Recommended ResourcesSebastian ThrunAndrew NgRana el Kaliouby"How Not To Be Wrong: The Power of Mathematical Thinking" by Jordan Ellenberg"Weapons of Math Destruction" by Cathy O'NeilHere are the codes for free eBook copies of Luis' book "Grokking Machine Learning": gmldcr-D659, gmldcr-2512, gmldcr-0752, gmldcr-30A2, gmldcr-01E8. Additionally, use the code poddcast19 to receive a 40% discount of all Manning products!

    Episode 46: From Building Recommendation Systems To Teaching Online Courses with Frank Kane

    Play Episode Listen Later Oct 29, 2020 47:46


    Show Notes(2:05) Frank reflected on his undergraduate experience studying Electrical Engineering at the University of Massachusetts - Dartmouth.(3:33) Frank commented on his experience working in the game industry after school.(6:28) Frank went over the opportunity to work as a software engineer at Amazon, where he contributed to the personalization system that recommends products to customers at a scale of tens of thousands of requests per second.(8:44) Frank brought up the challenges of building Amazon’s recommendation systems back in the early days.(10:14) Frank discussed how Amazon’s recommendations and content optimization technology evolved incrementally during his time as a Senior Manager.(12:05) Frank touched on the core engineering challenges during his time as a Senior Manager of Technology at IMDB.(14:19) Frank spoke about his proudest accomplishments at Amazon, both from the technical and the management perspectives.(18:19) Frank shared the story behind his professional transition into self-employment (check out his book “Self-Employment: Building an Internet Business of One”).(24:07) Frank shared a brief overview of his business (Sundog Software)'s virtual reality products.(25:15) Frank shared how he came to be an online instructor, discussed the pros/cons, and gave advice for aspiring ones.(29:38) Frank has created various courses that focus on Apache Spark, ranging from Python and Scala support to Spark Streaming capability.(31:34) Frank discussed how the Hadoop ecosystem has fallen out of favor (check out his popular Udemy courses titled “The Ultimate Hands-On Hadoop”).(33:20) Frank touched on ElasticSearch - an industry-standard open-source search engine (check out his Manning live videos on ElasticSearch 6 and ElasticSearch 7).(37:08) Frank provided his perspectives on the current landscape of recommendation systems research and applications.(42:17) Frank advised scientists and engineers on how to communicate with non-technical colleagues effectively.(43:25) Closing segment.His Contact InfoWebsiteLinkedInTwitterFacebookYouTubeHis Recommended ResourcesAmazon Leadership PrinciplesSundog Software“Self-Employment: Building an Internet Business of One”“Building Recommender Systems with Machine Learning and AI"Jose PortillaKirill EremenkoAndrew Ng"Lean Startup" by Eric Ries"Architecting Modern Data Platforms" by Jan Kunigk, Ian Buss, Paul Wilkinson, Lars GeorgeUse the codes below to get a discount from Frank's live video course on Manning called "Machine Learning, Data Science and Deep Learning with Python":mldldcr-4DB2mldldcr-9FE8mldldcr-EA35

    Episode 45: Teaching Artificial Intelligence with Amita Kapoor

    Play Episode Listen Later Oct 21, 2020 56:34


    Show Notes(2:02) Amita described her educational background, studying Electronics at universities back in the 90s. She also professed her love for Asimov’s writings.(3:33) Amita talked about her reason to pursue a path of an academic professor.(5:13) Amita discussed her Ph.D. titled "Modeling, Design, and Applications of Optical Amplifiers and Long Period Gatings” at the University of Dehli and Karlsruhe Institute of Technology(8:38) Amita shared her opinions on how the education of neural networks has evolved in the last 20 years of her teaching career - including the programming language shift from using Fortran and C++ to Python and the importance of learning computer networking and operating systems.(14:29) Amita discussed her research that combines the concepts of social network analysis and neural networks to model user behavior in society (read the full paper here).(17:57) Amita talked about the process of writing the TensorFlow 1.x Deep Learning Cookbook with Antonio Gulli.(21:08) Amita went over TensorFlow Machine Learning Projects, co-authored alongside Ankit Jain and Armando Fandango.(23:19) Amita dived into Hands-On Artificial Intelligence for IoT - which discusses different AI techniques to build smart IoT systems, covering practical case studies in personal & home devices, industrial applications, and smart cities.(27:50) Amita explained the improvements in TensorFlow 2.0 from its previous version, referring to her book Deep Learning with TensorFlow 2 and Keras - 2nd Edition in collaboration with Antonio Gulli and Sujit Pal.(31:33) Amita went over her experience participating in the NASA Centennial Space Robotics Challenge in 2017, in which her team finished in the top 20 out of more than 100 teams worldwide.(34:49) Amita reflected on her volunteering experience with a group of friends to build an Acute Myeloid Leukemia detection system that won an award for the Intel showcase in 2019.(38:21) Amita unpacked her blog post looking at COVID19 from a data science perspective.(40:26) Amita described her mentoring work at Neuromatch Academy, a non-profit online course in computational neuroscience.(44:19) Amita shared her opinion on the benefits of an online classroom versus an in-person classroom.(51:19) Amita expressed her thoughts on the tech and data community in New Dehli.(52:24) Amita shared her hobby of writing science fiction stories.(53:41) Closing segment.Her Contact InfoWebsiteGoogle ScholarLinkedInTwitterGithubHer Recommended Resources"TensorFlow 1.x Deep Learning Cookbook.""TensorFlow Machine Learning Projects""Hands-On Artificial Intelligence for IoT""Deep Learning with TensorFlow 2 and Keras - 2nd Edition."TensorFlow Strategy for Distributed TrainingNeuromatch Academy Course ContentComputational Neuroscience Coursera CourseAlan TuringJJ HopfieldGeoffrey Hinton“The Theory of Everything” by Stephen Hawking 

    Episode 44: Computer Systems, Machine Learning Security Research, and Women in Tech with Shreya Shankar

    Play Episode Listen Later Oct 12, 2020 64:53


    Show Notes(2:02) Shreya discussed her initial exposure to Computer Science and her favorite CS course on Advanced Topics in Operating Systems at Stanford.(4:07) Shreya emphasized the importance of distilling technical concepts to a non-technical audience, thanks to her experience as a section leader and teaching assistant for CS198.(6:26) Shreya shared the lack of representation in technical roles that keep women away from considering technology as a career path, and the initiative she was involved with at SHE++.(9:40) Shreya reflected on her software engineering internship experience at Facebook, working on Civic Engagement tools to help representatives connect with their constituents.(12:33) Shreya went over the anecdote of how she worked on Machine Learning Security research at Google Brain.(15:36) Shreya unpacked the paper “Adversarial Examples That Fool Both Computer Vision and Time-Limited Humans,” - where her team constructs adversarial examples that transfer computer vision models to the human visual system.(20:08) Shreya reflected on the lessons learned from her experience working with seasoned researchers at Google Brain.(23:31) Shreya gave her advice for engineers who are interested in multiple specializations.(25:34) Shreya provided resources on the fundamentals of computer systems.(27:15) Shreya explained her reason to work at an early-stage startup right after college (check out the blog post on her decision-making process).(28:41) Shreya was the first ML Engineer at Viaduct, a startup that develops end-to-end machine learning and data analytics platform to empower OEMs to manage, analyze, and utilize their connected vehicle data.(32:27) Shreya discussed two common misconceptions people have about the differences between machine learning in research and practice (read her reflection on one-year of making ML actually useful).(35:24) Shreya expanded on the organizational silo challenge that hinders collaboration between data scientists and software engineers while designing a machine learning product.(40:48) Shreya has been quite open about the challenge of recruiting female engineers, explaining that it is hard to sell women candidates when their alternatives are “conventionally sexy."(47:24) Shreya and a few others have developed and open-sourced GPT-3 Sandbox, a library that helps users get started with the GPT-3 API.(51:52) Shreya explained her prediction on why OpenAI can be the AWS of modeling.(54:24) Shreya shared the benefits of going to therapy to cope with mental illness challenges.(58:36) Closing segment.Her Contact InfoWebsiteTwitterLinkedInGitHubGoogle ScholarMediumHer Recommended ResourcesMartin Kleppmann’s “Designing Data-Intensive Applications”Stanford’s CS110 - “Principles of Computer Systems"Ada LovelaceWomen in AIBlack in AIQuoc LeUber Engineering BlogSteve Krug’s “Don’t Make Me Think"

    Episode 43: From Economics and Operations Management to Data Science with Francesca Lazzeri

    Play Episode Listen Later Sep 28, 2020 85:18


    Show Notes(2:37) Francesca discussed her educational background in Italy, studying Economics and Institutional Studies at LUISS Guido Carli University for her Master’s and then Economics and Technology Innovation at Sant’Anna University for her Ph.D. She also mentioned her transition to studying in the US at Harvard Business School.(7:43) Francesca shared the anecdote behind going to HBS to pursue a Postdoc Research Fellowship in Economics. She also revealed the differences in the educational approaches between Italy and the United States.(15:15) During her Postdoc, Francesca worked on multiple patent data-driven projects to investigate and measure the impact of external knowledge networks on companies’ competitiveness and innovation. She discussed a specific project that analyzed biotech innovation in Boston, San Diego, and San Francisco clusters using social media and citation data.(24:26) Francesca talked about her decision to join Microsoft as a data scientist in its Cloud and Enterprise division back in 2014, where she first worked on projects for clients from the energy and finance sectors.(30:00) Francesca discussed the two types of customers who seek Microsoft’s cloud solutions to solve their data problems and explained the learning curves she went through while interacting with them.(36:11) Francesca unpacked the Healthy Data Science Organization Framework - which is a portfolio of methodologies, technologies, resources that will assist organizations in becoming more data-driven (Read her InfoQ article “The Data Science Mindset: 6 Principles to Build Healthy Data-Driven Organizations”).(45:31) Francesca shared the challenges of building end-to-end machine learning applications that she has observed from Microsoft Azure AI’s clients.(49:56) Francesca walked through a typical day in her current leadership role at Microsoft’s Cloud AI Advocates team.(53:44) Francesca discussed the different components in a typical Azure deployment workflow (Read her post “Azure Machine Learning Deployment Workflow”).(58:44) Francesca explained Automated Machine Learning, a breakthrough from Microsoft Research division that is essentially a recommender system for machine learning pipelines.(01:03:50) Francesca went over model interpretability features within Azure AI (as part of the InterpretML package) and touched on Microsoft’s Responsible AI principles.(01:08:01) Francesca explained the differences between model fairness and model interpretability at both the training time and inference time (Check out the Fairlearn package).(01:12:11) Francesca is currently writing a book with Wiley called “Machine Learning for Time Series Forecasting with Python.”(01:14:39) Francesca shared her advice for undergraduate students looking to get into the field, judging from her experience being a mentor for Ph.D. and Postdoc students at institutions such as Harvard, MIT, and Columbia.(01:17:27) Francesca reasoned how her educational backgrounds in economics and operations management contribute to her success in a data science career(01:20:09) Closing segment.Her Contact InfoTwitterMediumLinkedInHer Recommended ResourcesPeople To FollowHilary MasonAndrew NgHannah WallachBook To ReadAn Introduction to Probability Theory and Its Applications (by William Feller)A Developer’s Introduction to Data ScienceVideo series on Data Science and Machine Learning on AzureVideo series on Data Science and Machine Learning on Azure GitHub repoAzure Machine LearningAzure Machine Learning DocumentationAzure Machine Learning ServiceThe Data Science LifecycleAlgorithm Cheat SheetHow to Select Machine Learning AlgorithmsAzure Machine Learning DesignerResponsible Machine LearningResponsible Machine LearningModel InterpretabilityInterpretML RepoInterpretML ToolkitInterpretML DocumentationFairlearn ServiceFairlearn DocumentationAutomated Machine LearningAutomated Machine LearningAuto ML FeaturizationAutoML Config Class

    Episode 42: Privacy-Preserving Natural Language Processing with Patricia Thaine

    Play Episode Listen Later Sep 11, 2020 55:11


    Show Notes(2:55) Patricia talked about his interest in learning languages and living in different cultures.(4:05) Patricia talked about her experience volunteering as a translator at the International Network of Street Papers.(5:00) Patricia studied Liberal Arts at John Abbott College, English Literature at Concordia University, and Computer Science and Linguistics at McGill University during her undergraduate years.(8:06) Patricia worked at McGill Language Development Lab as a Research Assistant, which studied how children learn different types of words and sentences.(9:15) Patricia described her graduate school experience at the University of Toronto, where she researched lost language decipherment and writing systems.(11:19) Patricia talked about MedStory, which is a text-oriented visual prototype built to support the complexity of medical narratives (spearheaded by Nicole Sultanum).(12:35) Patricia explained her research paper, “Vowel and Consonant Classification through Spectral Decomposition.”(15:29) Patricia unpacked her blog post, “Why is Privacy-Preserving NLP Important?”(19:02) Patricia dissected her paper “Privacy-Preserving Character Language Modelling” that proposes a method for calculating character bigram and trigram probabilities over sensitive data using homomorphic encryption.(21:13) Patricia wrote a two-part series called “Homomorphic Encryption for Beginners.”(22:21) Patricia unwrapped her paper “Efficient Evaluation of Activation Functions over Encrypted Data” that shows how to represent the value of any function over a defined and bounded interval, given encrypted input data, without needing to decrypt any intermediate values before obtaining the function’s output.(25:33) Patricia elaborated on her paper “Extracting Bark-Frequency Cepstral Coefficients from Encrypted Signals,” which claims that extracting spectral features from encrypted signals is the first step towards achieving secure end-to-end automatic speech recognition over encrypted data.(27:38) Patricia explained why privacy is an essential attribute for speech recognition applications.(29:53) Patricia discussed her comprehensive guide on “Perfectly Privacy-Preserving AI” which dives into the four pillars of perfectly privacy-preserving AI and outlines potential combinatorial solutions to satisfy all four pillars.(37:53) Patricia shared her take on the differences working in academic and commercial settings (she is the founder and CEO of Private AI).(40:50) Patricia talked about Private AI’s GALATEA Anonymization Suite, which anonymizes data at the source and encrypts them using quantum-safe cryptography.(45:05) Patricia emphasized the importance of talking to customers when building a commercial product.(46:58) Patricia shared her experience as a Postgraduate Affiliate at Vector Institute, which works with institutions, industry, startups, incubators, and accelerators to advance AI research and drive its application, adoption, and commercialization across Canada.(49:09) Patricia shared her advice for young researchers by going deep into at least two domains and combining the knowledge.(50:30) Patricia shared her excitement for privacy and NLP research in the upcoming years.(52:36) Closing segment.Her Contact InfoWebsiteTwitterLinkedInGoogle ScholarMediumGitHubHer Recommended ResourcesHomomorphic EncryptionSecure Multiparty ComputationFederated LearningDifferential PrivacyVector InstituteMILA Montreal InstituteAlberta Machine Intelligence InstituteReza Shokri (Assistant Professor at National University of Singapore)Parinaz Sobhani (Director of Machine Learning at Georgian Partners)Doina Precup (Associate Professor at McGill University)

    Episode 41: Effective Data Science with Eugene Yan

    Play Episode Listen Later Sep 3, 2020 94:55


    Show Notes(2:19) Eugene got his Bachelor’s degree in Psychology and Organizational Behavior from Singapore Management University, in which he did a senior thesis titled “Competition Improves Performance.”(3:29) Eugene’s first role out of school is an Investment Analyst position at Singapore’s Ministry of Trade & Industry.(4:18) Eugene then moved to a Data Analyst role at IBM, working on projects such as supply-chain dashboards, social media analytics, and anti-money laundering detection.(5:55) Eugene transitioned to an internal Data Scientist role at IBM, working on job forecasting and job recommendations.(9:03) Eugene shared the story of how he became a Data Scientist at Lazada Group, which was a small e-commerce startup back in 2015.(12:08) Eugene explained his decision to go back to school and pursued an online Master’s degree in Computer Science at Georgia Tech.(19:14) Eugene shared his career milestones, as displayed in his blog post reflecting on his journey from getting a degree in Psychology to leading data science at Lazada.(22:17) Eugene discussed the unique data science challenges while working at uCare.ai - a startup that aims to make healthcare more efficient and reduce costs.(25:29) Eugene revealed three useful tips to deliver great data science talks (read his blog post “How to Give a Kick-Ass Data Science Talk” for the details).(28:29) Eugene talked about his transition to become an Applied Scientist at Amazon - working on Amazon Kindle.(30:43) Eugene unpacked his post “Commando, Soldier, Police, and Your Career Choices” that provides an interesting metaphor to help guide career decisions.(33:43) Eugene went meta onto his writing process (read here) and note-taking strategy (read here).(39:01) Eugene shared the lessons learned from taking on responsibilities in hiring, mentoring, and stakeholder engagement in his second year at Lazada (read his blog post on the first 100 days as a Data Science Lead).(44:20) Eugene went in-depth into the engineering and cultural challenges throughout Alibaba Group’s acquisition of Lazada Group.(47:51) Eugene explained Alibaba’s playbook for the technical integration of their acquisitions and the super-apps phenomenon in Asia (check out a summary of his talk on Asia’s Tech Giants).(53:52) Eugene unpacked the values and essential aspects of Lazada’s data science team culture, as detailed in his post “Building a Strong Data Science Team Culture.”(57:44) Eugene summarized his thoughts on the topic of data science and agile/scrum development (Read his 3-part blog series: Part 1, Part 2, and Part 3).(01:03:18) Eugene was heavily involved with the development of product ranking, product recommendations, and product classification models in his first year at Lazada (check out slides to his talk “How Lazada Ranks Products”).(01:09:08) Eugene helped mentor and empower teams on multiple machine learning systems while acting as VP of Data Science at Lazada (check out slides to his talk “Data Science Challenges at Lazada”).(01:12:07) Eugene shared the case study of how uCare.ai developed a machine learning system for Southeast Asia’s largest healthcare group that estimates a patient’s total bill at the point of pre-admission.(01:14:06) Eugene summarized his 2-part series that exposes the challenges after model deployment and yields a practical guide to maintaining models in production.(01:19:04) Eugene discussed his early-career Product Classification project that uses a public Amazon dataset and builds two APIs for image classification & image search.(01:22:29) Eugene discussed his 2-part series that implements seven different models on the same Amazon dataset, from matrix factorization to graphs and NLP.(01:24:42) Closing segment.His Contact InfoWebsiteTwitterLinkedInGitHubHis Recommended ResourcesNiklas Luhmann (well-known German sociologist)Roam Research (note-taking application)MLflow (A platform for ML lifecycle management)Amazon Product Review Dataset (big data in JSON format)Andrej Karpathy (Read “The Unreasonable Effectiveness of RNNs” and “A Recipe For Training Neural Networks”)Jeremy Howard (Read the “Universal Language Model Fine-tuning for Text Classification paper)Hamel Hussain (Check out GitHub Actions and fastpages)“Introduction to Statistical Learning” (by Trevor Hastie and Rob Tibshirani)“The Pragmatic Programmer” (by Andy Hunt and Dave Thomas)applied-ml repositoryml-survey repository

    Episode 40: Biological Aging, Probabilistic Programming, and Private Machine Learning with Matthew McAteer

    Play Episode Listen Later Aug 22, 2020 77:55


    Show Notes(2:22) Matthew shared his childhood growing up interested in the field of biology.(5:29) Matthew described his undergraduate experience studying Cellular and Molecular Biology at Brown University. He dropped out for a year and a half to work at MIT and test out a few company ideas in the biotech space.(8:13) Matthew spent a decent amount of time in biological aging research after that, working at the Karp Lab at MIT and the Backsai Lab in Massachusetts General Hospital.(13:28) Matthew recalled the story of how he switched his pursuit to a career in Machine Learning.(17:14) Matthew commented on his experience as a Machine Learning Engineer freelancer on various projects in privacy and security, music analysis, and secure communications.(20:36) Matthew discussed the opportunity to work with Google as a contract software developer and shared valuable lessons from contributing to the TensorFlow Probability library for probabilistic reasoning and statistical analysis.(23:48) Matthew gave a quick overview of Bayesian Neural Networks (read his blog post for more details).(27:18) Matthew went over his contribution to the open-source community OpenMined, whose goal is to make the world more privacy-preserving by lowering the barrier-to-entry to private AI technologies.(32:29) Matthew worked on De-Moloch in late 2018, described to be "software that lets anyone easily run AI algorithms on sensitive data without it being personally identifiable" (read his blog post "Private ML Explained in 5 Levels of Complexity" for a complete description).(36:17) Matthew unpacked his post "Private ML Marketplaces," - which summarizes and discusses various approaches previously proposed in this space, such as smart contracts, data encryption/transformation/approximation, and federated learning.(39:45) Matthew shared his experience competing in the Pioneer Tournament.(42:19) Matthew shared brief advice on how to become a Machine Learning Engineer. For the full details, read his mega-post "Lessons from becoming an ML engineer in 12 months, without a CS or Math degree."(45:16) Matthew described his experience working as a Machine Learning Engineer at UnifyID, a startup that is building a revolutionary identity platform based on implicit passwordless authentication.(47:52) Matthew unpacked his research paper "Model Weight Theft with Just Noise Inputs: The Curious Case of the Petulant Attacker" at UnifyID. The paper explores the scenarios under which an attacker can steal the weights of a convolutional neural network whose architecture is already known.(51:55) Matthew is currently doing research with FOR.ai, a multi-disciplinary team of scientists and engineers who like researching for fun.(54:14) Matthew unpacked his research at FOR.ai, namely "Optimal Brain Damage" and "BitTensor: An Intermodel Intelligence Measure."(01:00:52) Matthew shared key takeaways from attending academic conferences such as ICML 2019 and NeurIPS 2019.(01:03:45) Matthew unpacked his 4-part series on ML Research interview that targets aspiring ML engineers, hiring managers/senior ML engineers, and people navigating ML research that don't want to lose sight of first principles.(01:07:09) Matthew unpacked his fantastic post called "Nitpicking ML Technical Debt" that breaks down relevant points of Google's famous paper on Hidden Technical Debt.(01:10:49) Matthew unpacked his well-researched list that examines the under-investigated fields in 10 academic domains ranging from computer science and biology to economics and philosophy.(01:14:41) Closing segment.His Contact InfoWebsiteTwitterLinkedInGitHubHis Recommended ResourcesVijay PandeDavid HaChip HuyenJohn Brockman's "This Idea Must Die"

    Episode 39: Serverless Machine Learning In Action with Carl Osipov

    Play Episode Listen Later Aug 7, 2020 89:11


    Show Notes(2:22) Carl talked about his early exposure to programming and his Bachelor’s degree in Computer Science at the University of Rochester in the late 90s.(5:12) Carl implemented his first fully connected, two-hidden-layer artificial neural network using the C programming language back in 2000 when using neural networks wasn’t nearly as cool as it is today.(8:00) Carl started his career as a software engineer at IBM, writing software for large-scale distributed systems and voice-dialog management system.(13:31) The first production machine learning system that Carl worked on is called Conversational Interaction Manager, which is a dialog management system for conversational mixed-initiative natural language applications. He brought up the challenges in DevOps and data quality.(20:05) The second production machine learning system that you worked on is called Smarter Campus, which is a project that enables staffing recommendations based on social networking, optimization, and text analytics.(27:16) Carl unpacked the evolution of his career at IBM, working on various leadership roles. In particular, he worked on IBM Bluemix, IBM’s cloud platform-as-a-service, with over 1 million registered users. He emphasized the importance of talking to customers and finding product-market fit.(33:01) Carl discussed his decision to pursue a Master’s degree in Computer Science at the University of Florida in the mid of his career.(35:24) Carl explained his research paper, which combines game theory and machine learning called “AmalgaCloud: Social Network Adaptation for Human and Computational Agent Team Formation.” The paper focuses on the relationship between network adaptation for candidate group participants and the performance of problem-solving groups.(40:50) Carl discussed his patent on learning ontologies for machine learning - which maps ontologies from data warehouses to computer systems.(47:00) Carl unpacked his 4-part blog series dated in 2016 that discusses server-less computing via tools such as Docker and Apache OpenWhisk.(52:58) Carl emphasized the importance of learning Docker to be productive as a Machine Learning practitioner.(55:02) Carl became a program manager at Google Cloud and helped manage the company’s efforts to democratize machine learning via the Advanced Solutions Lab in 2017.(59:07) Carl recalled his experience as an instructor at various machine learning boot camps.(01:01:44) Carl went over the growing popularity of semi-structured data, referring to his talk at Google’s 2018 Data Cloud Next event.(01:06:29) Currently, Carl is the CTO of CounterFactual AI, which works with various clients using tools such as PyTorch and AWS. He brought up an example of a food delivery application.(01:09:13) Carl went over his experience leading a workshop on Server-less Machine Learning with TensorFlow at the Reinforce AI Conference in Budapest last year.(01:10:52) Carl is writing a book with Manning called Server-less Machine Learning In Action. He explained that server-less tools help minimize the efforts to do MLOps.(01:13:47) Carl talked about the rise of PyTorch as a production-ready deep learning framework, as well as his preference for the PyTorch’s language design philosophy.(01:17:10) Carl shared his opinions on choosing different cloud platforms to host and run the server-less ML pipeline.(01:19:37) Carl described the data and tech community in Orlando, Florida.(01:21:53) Closing segment.His Contact InformationWebsiteLinkedInTwitterGitHubGoogle ScholarHis Recommended Resources“An Introduction to Natural Computation” by Dana BallardIBM’s semiconductor facility FABIBM Bluemix“Pattern Recognition and Machine Learning” by Christopher BishopDockerPyTorchPyTorch LightningJurgen Schmidhuber (the father of LSTM)Solomon Hykes (Founder, CTO, and chief architect of Docker)Dana Ballard (professor of Computer Science at UT-Austin)Gang of Four Design Patterns (engineering book with object-oriented design theory and practice)Serverless Machine Learning In ActionCheck out the book at this link: https://www.manning.com/books/serverless-machine-learning-in-action?a_aid=khanhnamle1994&a_bid=fa913283Here are the 5 free Ebook codes: smldcr-0BE0, smldcr-9F05, smldcr-F807, smldcr-CBB2, smldcr-52D1Here is a 40% discount code: poddcast19

    Episode 38: Designing For Analytics with Brian O'Neill

    Play Episode Listen Later Jul 27, 2020 75:29


    Show Notes(2:40) Brian discussed his career as a musician and his mission as a consultant to bring design principles into the analytics world.(5:25) Brian talked about his career inception as a UX designer and his interest in human-centered design.(9:48) Brian shared the backstory behind starting Designing for Analytics and his advice for anyone interested in becoming a consultant (hint: finding the minimum viable audience for your craft!).(20:44) Brian shared the common problems that his clients ask him to solve - citing that many of the solutions in engineering-driven organizations are “Technically Right, Effectively Wrong” (listen to Brian’s podcast with David Stephenson).(27:20) Brian explained why data product design goes well beyond user interfaces and helps define what is required to enable the desired user and business outcomes, referring to his post “Does your data product enable surgery or healing?”(33:14) Brian revealed the tactical tips for designing an effective prototype for data products, as shared in his post “Designing MVPs for Data Products and Decision Support Tools.”(40:31) Brian talked about the importance of using human-centered design to measure meaningful engagement in the context of data products, as shared in his post “Why Low Engagement May Not be the Problem with Your Data Product or Analytics Service” (Hint: Think about the last mile and use design to make deliberate choices to improve user engagement).(47:46) Brian unpacked the design framework CED (which stands for Conclusion, Evidence, and Data), which helps build customer trust, engagement, and indispensability around advanced analytics.(54:55) Brian shared his take on how to structure a quad team, including software engineers, UX designers, data scientists, and product managers to build machine learning-powered products.(01:03:25) Brian emphasized the importance of trust in modern data products, after countless conversations with leaders in his podcast Experiencing Data.(01:06:34) Brian unveiled his seminar called Designing Human-Centered Data Products for data scientists, technical product managers, and analytics practitioners.(01:09:47) Closing segment.His Contact InfoDesigning For AnalyticsTwitterLinkedInExperiencing Data PodcastInsights NewsletterHis Recommended ResourcesSeth Godin’s podcast AkimboMinimum Viable ProductWizard of Oz TestingCED FrameworkChris DoScott Berkun (his book “How Design Makes The World”)Juhan Sonin (Involution Studios)Amanda Cox (NYT’s The Upshot)“Good Charts” (by Scott Berinato)“Change By Design” (by Tim Brown)“Infonomics” (by Douglas Laney)“Competing In The Age of AI” (by Marco Iansiti and Karim Lakhani)

    Episode 37: Machine Learning In Production with Luigi Patruno

    Play Episode Listen Later Jul 20, 2020 102:14


    Show Notes(2:19) Luigi got his Bachelor’s in Mathematics and Master’s in Computer Science from Fordham University, with a break working as a Data Analyst in between.(5:41) Luigi worked as a Research Engineer at Fordham's Wireless Sensor Data Mining Lab for a year during his Master’s program and got exposed to Machine Learning.(9:13) Luigi’s first role out of graduate school is a Data Engineering position at Namely, a Human Resources platform for thousands of mid-sized companies.(14:33) Luigi then worked as a Machine Learning Engineer at CTRL-Labs - a startup (acquired by Facebook) pioneering the development of non-invasive neural interfaces that reimagine how humans and machines collaborate.(20:45) Luigi discusses the skills he picked up during his transition from Data Engineering to Machine Learning Engineering, such as data analysis, data visualization, dimensionality reduction, and domain expertise.(25:38) Luigi went over his time teaching graduate courses in Applied Statistics & Probability and Big Data Programming at Fordham’s Department of Computer Science.(28:37) Luigi talked about his next role working as a Data Scientist at 2U - an edTech SaaS platform providing schools with the comprehensive operating infrastructure they need to attract, enroll, educate, support, and graduate students globally.(31:12) Luigi emphasized the importance of being good at data science and picking up skills from other functional domains for anyone looking into management roles.(33:47) Luigi shared brief thoughts on the role of ed-tech in the current environment with remote education.(35:47) Luigi unpacked his blog post called How I Hire Data Scientists that shares advice for both the hiring managers and the job applicants.(42:05) Luigi shared his anecdotal journey of starting ML In Production - which provides content on the best practices of doing machine learning in production. Check out this article for more detail!(47:16) Luigi discussed the nuts and bolts of setting up the weekly newsletter for his website.(51:21) Luigi unpacked the 4-part series "Docker for Machine Learning” that discusses the benefits of using Docker with machine learning, how to build custom Docker images, how to perform batch inference using Docker containers, and how to perform online inference using Docker and Flask REST API.(54:26) Luigi’s next post, "Batch Inference vs. Online Inference," discusses the differences between using batch inference or online inference for model serving.(56:51) Luigi’s next post, "Storing Metadata from ML Experiments," reveals the importance of storing metadata during the machine learning process as well as the types of metadata to capture.(01:00:49) Luigi’s following post "How Data Leakage Impacts ML Models" goes over the issues of data leakage, which occurs when data used at training time is unavailable at inference time.(01:04:14) Luigi unpacked his 6-part series that first introduces Kubernetes and then goes deeper into its components, including Pods, Jobs, CronJobs, Deployment, and Services.(01:07:06) Luigi reflected on his talk “Productionizing ML Models at scale with Kubernetes” at the TWIML conference last year.(01:10:50) Luigi dug into his popular post "The Ultimate Guide to Model Retraining," which covers the problem of model drift as well as the necessary steps to retrain models already in production.(01:14:36) Luigi laid out the benefits of using AWS SageMaker for model deployment. Check out his concise description of SageMaker’s architecture as well as his video tutorial on how to train scikit-learn models on SageMaker.(01:17:50) Luigi unpacked his multi-part series on model deployment. So far, he has covered deployment in the machine learning context, software interfaces, batch inference, online inference, model registries, test-driven development, and A/B testing.(01:22:12) Luigi encouraged every software engineer to learn about running ML systems in production, given the gradual shift to Software 2.0, as indicated in his post "Machine Learning is Forcing Software Development to Evolve."(01:26:48) Luigi reveals what differentiates successful industry ML projects from unsuccessful ones, based on his interview series with other ML practitioners. Hint: (1) don’t focus on the hype, instead focus on the business outcomes + (2) start small.(01:30:10) Luigi distinguished the skills required for the three roles: data engineers, data scientists, and machine learning engineers.(01:33:46) Luigi shared his opinions on the data science community in New York City.(01:35:41) Closing segment.His Contact InfoWebsiteTwitterLinkedInHis Recommended ResourcesMLinProduction.comMonitor! Stop Being a Blind Data Scientist (by Ori Cohen)Andrej KarpathyLex FridmanXavier Amatriain“Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow” by Aurelien GeronA New Course From LuigiLuigi just launched his first online course, Build, Deploy, and Monitor Machine Learning Models with Amazon SageMaker! I had a look at the course content, and I’m convinced that the course will be super valuable to any ML engineer or data scientist who wants to level up and learn how to productionize their machine learning models.You can take the course on your own, but Luigi is also teaming up with TWiML to offer a version with virtual Study Group sessions for people who want a more interactive experience. Right now, Luigi is offering an early bird discount on the course until August 1st!I know the course will be precious for a lot of you within my community, so Luigi created a coupon code DATACAST to save an additional 10% off the course!Head over to the Teachable course page to learn more about AmazonSageMaker and take advantage of the discount!

    Episode 36: Machine Learning Bookcamp with Alexey Grigorev

    Play Episode Listen Later Jul 3, 2020 65:37


    Show Notes(2:00) Alexey studied Information Systems and Technologies from a local university in his hometown in eastern Russia.(4:54) Alexey commented on his experience working as a Java developer in the first three years after college in Russia and Poland, along with his initial exposure to Machine Learning thanks to Coursera.(7:55) Alexey talked about his decision to pursue the IT4BI Master Program specializing in Large-Scale Business Intelligence in 2013.(9:42) Alexey discussed his time working as a Research Assistant on Apache Flink at the DIMA Group at TU Berlin.(12:28) Alexey’s Master Thesis is called Semantification of Identifiers in Mathematics for Better Math Information Retrieval, which was later presented at the SIGIR conference on R&D in Information Retrieval in 2016.(14:35) Alexey discussed his first job as a Data Scientist at Searchmetrics - working on projects to help content marketers improve SEO ranking for their articles.(18:54) Alexey’s next role was with the ad-tech company Simplaex. There, he designed, developed, and maintained the ML infrastructure for processing 3+ billion events per day with 100+ million unique daily users - working with tools like Spark for data engineering tasks.(22:17) Alexey reflected on his journey participating in Kaggle competitions.(25:35) Alexey also participated in other competitions at academic conferences: winning 2nd place at the Web Search and Data Mining 2017 challenge on Vandalism Detection and winning 1st place at the NIPS 2017 challenge on Ad Placement.(29:59) Alexey authored his first book called Mastering Java for Data Science, which teaches readers how to create data science applications with Java.(31:40) Alexey then transitioned to a Data Scientist role at OLX Group, a global marketplace for online classified advertisements.(33:23) Alexey explained the ML system that detects duplicates of images submitted to the OLX marketplace, which he presented at PyData Berlin 2019. Read his two-part blog series: The first post presents a two-step framework for duplicate detection, and the second post explains how his team served and deployed this framework at scale.(38:12) Alexey was recently involved in building an infrastructure for serving image models at OLX. Read his two-part blog series on this evolution of image model serving at OLX, including the transition from AWS SageMaker to Kubernetes for model deployment, as well as the utilization of AWS Athena and MXNet for design simplification.(42:39) Alexey is in the process of writing a technical book called Machine Learning Bookcamp - which encourages readers to learn machine learning by doing projects.(46:17) Alexey discussed common struggles during data science interviews, referring to his talk on Getting a Data Science Job.(48:32) Alexey has put together a neat GitHub page that includes both theoretical and technical questions for people who are preparing for interviews.(52:19) Alexey extrapolated on the steps needed to become a better data scientist, in conjunction to his LinkedIn post a while back.(56:40) Alexey gave his advice for software engineers looking to transition into data science.(58:32) Alexey shared his opinion on the data science community in Berlin.(01:01:53) Closing segment.His Contact InfoWebsiteTwitterLinkedInGitHubKaggleQuoraGoogle ScholarMediumHis Recommended ResourcesApache FlinkKubeflowData Science Interviews GitHub RepoPyData BerlinBerlin BuzzwordsAndrew NgDesigning Data-Intensive Applications by Martin KleppmannMachine Learning BookcampPermanent 40$ discount code: poddcast195 free eBook codes (each good for one sample of the book): mlbdrt-D452, mlbdrt-5922, mlbdrt-2C4D, mlbdrt-3034, mlbdrt-1DD1

    Episode 35: Data Science For Food Discovery with Ankit Jain

    Play Episode Listen Later Jun 20, 2020 44:50


    Show Notes(2:27) Ankit studied Electrical Engineering with a focus on Communication and Signal Processing at the Indian Institute of Technology, Bombay.(3:27) Ankit then worked for three years as a Senior Field Engineer at Schlumberger, an international oilfield services company.(4:23) Ankit then went to the US to pursue a Masters in Financial Engineering from the Walter Hass School of Business at UC Berkeley.(6:13) Ankit had an opportunity to intern as a data scientist at Facebook during his Masters and worked on detecting spam for Facebook pages.(8:27) Ankit worked full-time as a Quantitative Finance Analyst at Bank of America after finishing his degree, with projects such as building models to identify risk in bank portfolio and analyzing relevance opportunities for strategic investment.(9:46) Ankit discussed his transition to a Data Scientist role at ClearSlide, a B2B platform for Sales Enablement + Engagement.(11:32) Ankit discussed his work on sales forecasting algorithms at ClearSlide.(15:06) In 2015, Ankit moved to Bangalore to become the Head of Data Science and Analytics at Ruunr, a B2B platform that offers hyper-local logistics services that partners with merchants in India.(16:18) Ankit unpacked his thorough post “How Food Delivery Can Be a Sustainable Business” that reflects his experience at Ruunr.(18:35) Ankit talked about the similarities and differences of tech culture in Bangalore and San Francisco.(19:39) Ankit came back to the US and started working as a Data Scientist at Uber in early 2017.(20:34) Ankit discussed his work at Uber on user-level forecasting.(23:12) Ankit talked about the different types of problems that researchers at Uber AI Labs work on.(24:49) Ankit unpacked his in-depth technical post on Uber’s Engineering blog “Food Discovery with Uber Eats: Using Graph Learning to Power Recommendations” — including graph neural networks for food recommendations, the design of the data and training pipeline, and ways to incorporate more data for further improvement.(28:55) Ankit discussed the challenges with building the Uber Eats recommendation system in production.(32:15) Ankit has written a technical book called TensorFlow Machine Learning Projects — which teaches how to exploit the benefits (simplicity, efficiency, and flexibility) of using TensorFlow in various real-world projects.(34:43) Ankit gave his two cents on the battle of frameworks between TensorFlow and PyTorch.(36:26) Ankit shared his advice for academics looking to work in the industry: building end-to-end projects, learning how to build scalable pipelines, and keeping up with important research topics.(38:33) Ankit reflected on the benefits of his electrical engineering and financial analysis education towards his career in data science.(40:11) Closing segment.His Contact InfoLinkedInTwitterGitHubQuoraHis Recommended ResourcesGraphSAGEMeta-Graph: Few-Shot Link Prediction via Meta-LearningAnkit’s book "TensorFlow Machine Learning Projects” published with PacktAndrew NgGeoffrey HintonJeff Dean"Elements of Statistical Learning" by Trevor Hastie, Robert Tibshirani, and Jerome Friedman

    Episode 34: Deep Learning Generalization, Representation, and Abstraction with Ari Morcos

    Play Episode Listen Later Jun 14, 2020 97:03


    Show Notes(2:32) Ari discussed his undergraduate studying Physiology and Neuroscience at UC San Diego, while doing neuroscience research on adult neurogenesis at the Gage Lab.(4:39) Ari discussed his decision to pursue a Ph.D. in Neurobiology at Harvard after college and extracted the importance of communication in research, thanks to his advisor Chris Harvey.(7:16) Ari explained his Ph.D. thesis titled “Population dynamics in parietal cortex during evidence accumulation for decision-making” - in which he developed methods to understand how neuronal circuits perform the computations necessary for complex behavior.(12:59) Ari talked about his process of learning machine learning and using that to analyze massive neuroscience datasets in his research.(15:22) Ari recounted attending NIPS 2015 and serendipitously meeting people from DeepMind, which he lated joined as a Research Scientist in their London office.(18:59) Ari’s research focuses on the generalization of neural networks, and shared his work called "On the Importance of Single Directions for Generalization” presented at ICLR 2018 (inspired by Chiyuan Zhang’s paper and Quoc Le’s paper previously).(28:51) Ari explained the differences between generalizing networks and memorizing networks, citing the results from his work "Insights on Representational Similarity in Neural Networks with Canonical Correlation” with Maithra Raghu and Samy Bengio presented at NeurIPS 2018 (Read Maithra’s paper on SVCCA that inspired it).(35:16) Another topic that Ari focuses on is representation learning and abstraction for intelligent systems. His team at DeepMind proposes a dataset and a challenge designed to probe abstract reasoning, as explained in “Measuring Abstract Reasoning in Neural Networks" presented at ICML 2018 (learn more about the IQ test Raven’s Progressive Matrices and take the challenge here).(42:21) An extension from the work above is "Learning to Make Analogies by Contrasting Abstract Relational Structure" - presented at ICLR 2019. With the same authors (led by Felix Hill along with David Barrett, Adam Santoro, Tim Lillicrap), Ari showed that while architecture choice can influence generalization performance, the choice of data and the manner in which it is presented to the model is even more critical.(48:18) Ari discussed "Neural Scene Representation and Rendering” (led by Ali Eslami and Danilo Rezende) that introduces Generative Query Network (GQN), a framework within which machines learn to represent scenes using only their own sensors (watch the video and check out the data).(55:09) Ari explained the findings in "Analyzing Biological and Artificial Neural Networks: Challenges with Opportunities for Synergy?” published at the Current Opinion in Neurobiology (joint work with David Barrett and Jakob Macke).(57:04) Ari shared the properties of pruning algorithms that influence stability and generalization, as claimed in “The Generalization-Stability Tradeoff in Neural Network Pruning” led by Brian Bartoldson.(01:00:56) Ari went over the generalization of lottery tickets in neural networks, which is inspired by the lottery ticket hypothesis from Jonathan Frankle and Michael Carbin at MIT. The two papers mentioned are collaboration with Haonan Yu, Yuandong Tian, Michela Paganini, and Sergey Edunov (Check out his talk at REWORK Deep Learning Summit in Montreal 2019).(01:09:00) Ari investigated "Training BatchNorm and Only BatchNorm” which looks at the performance of neural networks when trained only with the Batch Normalization parameters (joint work with Jonathan Frankle and David Schwab).(01:12:12) Ari mentioned "The Early Phase of Neural Network Training” (presented at ICML 2020) that uses the lottery ticket framework to rigorously examine the early part of the training (joint work with Jonathan Frankle and David Schwab). (01:16:25) Ari discussed at length “Representation Learning Through Latent Canonicalizations" (presented at ICLR 2020). This work seeks to learn representations in which semantically meaningful factors of variation (like color or shape) can be independently manipulated by learned linear transformations in latent space, termed “latent canonicalizes” (joint work with Or Litany, Srinath Sridhar, Leonidas Guibas, and Judy Hoffman).(01:22:15) Ari summarized "Selectivity Considered Harmful: Evaluating the Causal Impact of Class Selectivity in DNNs" - which investigates the causal impact of class selectivity on network function (led by Matthew Leavitt).(01:25:26) Ari reflected on his career and shared advice for individuals who want to make a dent in AI research.(01:28:10) Ari shared his excitement on self-supervised learning, which addresses the need of neural networks to require expensive labeled data.(01:29:47) Closing segment.His Contact InformationWebsiteGoogle ScholarLinkedInTwitterGitHubHis Recommended Resources“Understanding Deep Learning Requires Rethinking Generalization” by Chiyuan Zhang“Singular Vector Canonical Correlation Analysis for Deep Learning Dynamics and Interpretability” by Maithra RaghuRaven’s Progressive Matrices IQ test"The Lottery Ticket Hypothesis” by Jonathan Frankle and Michael Carbin (Open-Source Framework)“Random Features for Large-Scale Kernel Machines” by Ali Rahimi and Ben Recht (NIPS 2017 Test Of Time Award)“beta-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework” by DeepMindSamy Bengio (Research Scientist at Google AI)Aleksander Madry (Professor of Computer Science at MIT)Jason Yosinski (Founding Member of Uber AI Labs)“The Idea Factory: Bell Labs and The Great Age of American Innovation" by Jon Gertner

    Episode 33: Domain Randomization in Robotics with Josh Tobin

    Play Episode Listen Later Jun 3, 2020 53:47


    Show Notes:(2:02) Josh studied Mathematics at Columbia University during his undergraduate and explained why he was not set out for a career as a mathematician.(3:55) Josh then worked for two years as a Management Consultant at McKinsey.(6:05) Josh explained his decision to go back to graduate school and pursue a Ph.D. in Mathematics at UC Berkeley.(7:23) Josh shared the anecdote of taking a robotics class with professor Pieter Abbeel and switching to a Ph.D. in the Computer Science department at UC Berkeley.(8:50) Josh described the period where he learned programming to make the transition from Math to Computer Science.(10:46) Josh talked about the opportunity to collaborate and then work full-time as a Research Scientist at OpenAI - all during his Ph.D.(12:40) Josh discussed the sim2real problem, as well as the experiments conducted in his first major work "Transfer from Simulation to Real World through Learning Deep Inverse Dynamics Model".(17:43) Josh discussed his paper "Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real World", which has been cited more than 600 times up until now.(20:51) Josh unpacked the OpenAI’s robotics system that was trained entirely in simulation and deployed on a physical robot, which can learn a new task after seeing it done once (Read the blog post “Robots That Learn” and watch the corresponding video).(24:01) Josh went over his work on Hindsight Experience Replay - a novel technique that can deal with sparse and binary rewards in Reinforcement Learning (Read the blog post “Generalizing From Simulation").(28:41) Josh talked about the paper "Domain Randomization and Generative Models for Robotic Grasping”, which (1) explores a novel data generation pipeline for training a deep neural network to perform grasp planning that applies the idea of domain randomization to object synthesis; and (2) proposes an autoregressive grasp planning model that maps sensor inputs of a scene to a probability distribution over possible grasps.(32:27) Josh unpacked the design of OpenAI's Dactyl - a reinforcement learning system that can manipulate objects using a Shadow Dexterous Hand (Read the paper “Learning Dexterous In-Hand Manipulation” and watch the corresponding video).(35:31) Josh reflected on his time at OpenAI.(36:05) Josh investigated his most recent work called “Geometry-Aware Neural Rendering” - which tackles the neural rendering problem of understanding the 3D structure of the world implicitly.(28:21) Check out Josh's talk "Synthetic Data Will Help Computer Vision Make the Jump to the Real World" at the 2018 LDV Vision Summit in New York.(28:55) Josh summarized the mental decision tree to debug and improve the performance of neural networks, as a reference to his talk "Troubleshooting Deep Neural Networks” at Reinforce Conf 2019 in Budapest.(41:25) Josh discussed the limitations of domain randomization and what the solutions could look like, as a reference to his talk "Beyond Domain Randomization” at the 2019 Sim2Real workshop in Freiburg.(44:52) Josh emphasized the importance of working on the right problems and focusing on the core principles in machine learning for junior researchers who want to make a dent in the AI research community.(48:30) Josh is a co-organizer of Full-Stack Deep Learning, a training program for engineers to learn about production-ready deep learning.(50:40) Closing segment.His Contact Information:WebsiteLinkedInTwitterGitHubGoogle ScholarHis Recommended Resources:Full-Stack Deep LearningPieter AbbeelIlya SutskeverLukas Biewald“Thinking Fast and Slow” by Daniel Kahneman

    Episode 32: Economics, Data For Good and AI Research with Sara Hooker

    Play Episode Listen Later May 21, 2020 93:38


    Show Notes:(2:20) Sara shared her childhood growing up in Africa.(4:05) Sara talked about her undergraduate experience at Carleton College studying Economics and International Relations.(9:07) Sara discussed her first job working as an Economics Analyst at Compass Lexecon in the Bay Area.(12:20) Sara then joined Udemy as a data analyst, then transitioned to the engineering team to work on spam detection and recommendation algorithms.(14:58) Sara dig deep into the “hustling period” of her career and how she brute-forced her way to grow as an engineer.(17:24) Sara founded Delta Analytics - a local Bay Area non-profit community of data scientists, engineers, and economists in 2014 that believes in using data for good.(20:53) Sara shared Delta’s collaboration with Eneza Education to empower students to access quizzes by mobile texting in Kenya (check out her presentation at the ODSC West 2016).(25:16) Sara shared Delta’s partnership with Rainforest Connection to identify illegal de-forestation using steamed audio from the rainforest (check out her presentation at MLconf Seattle 2017).(28:22) Sara unpacked her blog post Why “data for good” lacks precision, in which she described 4 key criteria frequently used to qualify an initiative as “data for good” and discussed some open challenges associated with each.(36:34) Sara unpacked her blog post Slow learning, in which she revealed her journey to get accepted into the AI Residency program at Google AI.(41:03) Sara discussed her initial research interest on model interpretability for deep neural networks and her work done at Google called The (Un)reliability of Saliency Methods - which argues that saliency methods are not reliable enough to explain model prediction.(45:55) Sara pushed the research above further with A Benchmark for Interpretability Methods in Deep Neural Networks, which proposes an empirical measure of the approximate accuracy of feature importance estimates in deep neural networks called RemOve And Retrain.(48:46) Sara explained why model interpretability is not always required (check out her talks at PyBay 2018, REWORK Toronto 2018, and REWORK San Francisco 2019).(52:10) Sara explained the typical measurements of model reliability and the limitations of them, such as localization methods and points of failure.(59:04) Sara explained why model compression is an interesting research direction and her work The State of Sparsity in Deep Neural Networks - which highlights the need for large-scale benchmarks in the field of model compression.(01:02:49) Sara discussed her paper Selective Brain Damage: Measuring the Disparate Impact of Model Pruning - which explores the impact of pruning techniques for neural networks trained for computer vision tasks. Check out the paper website!(01:05:08) Sara shared her future research directions on efficient pruning, sparse network training, and local gradient updates.(01:06:56) Sara explained the premise behind her talk Gradual Learning at the Future of Finance Summit in 2019, in which she shared the three fundamental approaches to machine learning impact.(01:12:20) Sara described the AI community in Africa as well as the issues the community is currently facing: both from the investment landscape and the infrastructure ecosystem.(01:18:00) Sara and her brother recently started a podcast called Underrated ML which pitches the underrated ideas in machine learning.(01:20:15) Sara reflected how her background in economics influences her career outlook in machine learning.(01:25:42) Sara reflected on the differences between applied ML and research ML, and shared her advice for people contemplating between these career paths.(01:29:49) Closing segment.Her Contact Information:WebsiteLinkedInGitHubGoogle ScholarTwitterMediumHer Recommended Resources :Deep Learning IndabaSoutheast Asia Machine Learning SchoolMILA - AI For HumanityWhy “data for good” lacks precision (Sara's take on "Data for Good" initiatives)Slow learning (Sara's journey to Google AI)fast.ai Sanity Check for Saliency Maps by Julius Adebayo et al.Focal Loss for Dense Object Detection by Tsung-Yi Lin et al.MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications by Andrew Howard et al.Underrated ML (Sara’s new podcast)Dumitru Erhan (Research Scientist at Google AI)Samy Bengio (Research Scientist at Google AI)Andrea Frome (Ex-Research Engineer at Google AI)Elements of Statistical Learning by Trevor Hastie, Robert Tibshirani, and Jerome Friedman

    Episode 31: From Quantum Computing to Epidemic Modeling with Colleen Farrelly

    Play Episode Listen Later Apr 18, 2020 67:40


    Show Notes:(1:58) Colleen gave a brief overview of her professional background and her path to data science.(3:27) Colleen explained how her background in medicine and social science contributes to her success as a data scientist working in different domains.(5:20) Colleen share her thoughts on how data science varies by sector.(7:02) Referring to her consulting company Staticlysm LLC, Colleen discussed a current medical technology project that she is working on.(8:19) Colleen shared applications of quantum machine learning in the wild, referring to her work at Quantopo - where she is the cofounder and chief mathematician.(12:33) Colleen discussed a new project leveraging quantum and quantum-inspired algorithms for nuclear reactor optimization at Quantopo.(15:17) Colleen gave advice for data scientists who want to start a business and get into consulting.(16:58) Colleen discussed topological data analysis in machine learning.(23:02) Colleen discussed epidemic modeling and engaging foreign aid organizations, given her experience with the Ebola outbreak.(26:01) Colleen discussed the different approaches to model the spread of diseases in epidemics, which follow a differential equation framework.(30:22) Colleen explained why buy-in and corporation from those in power are critical to combat epidemics.(32:20) Colleen explained the difference between the current Coronavirus pandemic and the previous Ebola epidemic (note that this conversation is recorded in mid-March 2020, so information about COVID19 may have dated).(37:06) Colleen shared resources to get up-skilled in data science.(39:40) Colleen talked about the benefits of writing on Quora, where she has written more than 13,000 answers.(41:31) Colleen shared the traits of an excellent technical communicator and/or data translator.(43:26) Colleen shared her process of writing a technical book that focuses on the use-cases of topology, geometry, and graph theory in machine learning and data science.(50:32) Colleen talked about the growth of the data science community in Miami.(54:30) Colleen discussed her involvement in data science within the African sectors.(57:55) Colleen shared her thoughts on how the data science field will evolve in the next few years: the access to big data platforms, the dominance of Python, and the rise of quantum computing.(01:01:12) Closing segment.Her Contact Info:LinkedInQuoraResearchGateKDNuggetsHer Recommended Resources: IBM Quantum ComputingXanadu (Quantum Computing Hardware)D-Wave Systems (Quantum Computing Hardware)Ayasdi (Topological Data Analysis-Focused Startup)The SIR Model for Spread of DiseaseGoogle ScholararXivLinkedIn LearningCourseraGenetic Algorithms and Adaptation paper by John HollandRandom Forests paper by Leo BreimanAndrew Ng (Founder of Coursera, DeepLearning.AI, and Landing.AI)Dover Series on Mathematics

    Episode 30: Data Science Evangelism with Parul Pandey

    Play Episode Listen Later Mar 27, 2020 74:49


    Show Notes:(2:12) Parul talked about her educational background, studying Electrical Engineering at the National Institute of Technology, Hamirpur.(3:18) Parul worked as a Business Analyst at Tata Power India for 7 years.(4:29) Parul talked about her initial interests in writing about data science and machine learning on Medium.(6:30) Parul discussed her first blog series “A Guide to Machine Learning in R for Beginners” - which covers the Fundamentals of ML, Intro to R, Distributions and EDA in R, Linear Regression, Logistic Regression, and Decision Trees.(8:02) Reference to her articles on data visualization, Parul talked about matplotlib, seaborn, and plotly as the main visualization libraries she practices, in addition to Tableau for building dashboard.(10:11) Parul shared her thoughts on the state of Machine Learning interpretability, in reference to her articles on this topic.(13:54) Parul discussed the advantages of using Jupyter Lab over Jupyter Notebook.(17:30) Parul discussed the common challenges of bringing recommendation systems from prototype into production (Read her two articles about recommendation systems: (1) an overview of different approaches and (2) an overview of the process of designing and building a recommendation system pipeline)(21:00) Parul went in depth into her NLP project called "Building a Simple Chatbot from Scratch in Python (using NLTK).”(23:26) Parul continued this chatbot project with a 2-part series on building a conversational chatbot with Rasa stack and Python and deploying it on Slack.(28:15) Parul went over her Satellite Imagery Analysis with Python piece, which examines the vegetation cover of a region with the help of satellite data.(32:22) Parul talked about the process of Recreating Gapminder in Tableau: A Humble Tribute to Hans Rosling.(35:17) Parul discussed her project Music Genre Classification, which shows how to analyze an audio/music signal in Python.(39:20) Parul went over her tutorials on Computer Vision: (1) Face Detection with Python using OpenCV and (2) Image Segmentation with Python’s scikit-image module.(42:01) Parul unpacked her tutorial "Predicting the Future with Facebook’s Prophet” - a forecasting model to predict the number of views for her Medium articles.(44:58) Parul have been working as a Data Science Evangelist at H2O.AI since July 2019.(47:04) Parul described Flow - H2O's web-based interface (Read her tutorial here).(49:23) Parul described Driverless AI - H2O’s product that automates the challenging and repetitive tasks in applied data science (Read her tutorial here).(52:39) Parul described AutoML - H2O's automation of the end-to-end process of applying ML to real-world problems (Read her tutorial here).(57:07) Parul shared her secret sauce for effective data visualization and storytelling, as illustrated in her analysis of the 2019 Kaggle Survey to figure out women’s representation in machine learning and data science.(01:02:02) Parul described the data science community in Hyderabad, from her lens as an organizer for the Hyderabad Chapter of the Women in Machine Learning and Data Science.(01:05:45) Parul was recognized as a LinkedIn’s Top Voices 2019 in the Software Development category.(01:10:30) Closing segment.Her Contact Info:MediumGitHubTwitterLinkedInWebsiteKaggleHer Recommended Resources:Interpretable Machine Learning post"Interpretable Machine Learning: A Guide for Making Black Box Models Explainable" by Chris Molnar“Towards A Rigorous Science of Interpretable Machine Learning” by Finale Doshi-Velez and Been KimParul’s Compilation of Data Visualization articlesParul’s Programming with Python articlesWomen in Machine Learning and Data ScienceRachel ThomasAndreas MuellerHuggingFace“Factfulness” by Hans Rosling

    Episode 29: From Bioinformatics to Natural Language Processing with Leonard Apeltsin

    Play Episode Listen Later Mar 13, 2020 77:40


    Show Notes:(2:18) Leonard discussed his undergraduate experience at Carnegie Mellon - where he studied Biology and Computer Science.(5:10) Leonard decided to pursue a Ph.D. in Bioinformatics at the University of California - San Francisco.(6:27) Leonard described his Ph.D. research that focused on finding hidden patterns in genetically-linked diseases.(9:42) Leonard went deep into clustering algorithms (Markov Clustering and Louvain) and their applications such as protein and news article similarity.(13:21) Leonard shared his story of starting a data science consultancy with various client startups.(17:58) Leonard discussed the interesting consulting projects that he worked on: from detecting plagiarism to predicting bill insurance.(22:04) Leonard shared practical tips to learn technical concepts.(23:23) Leonard reflected on his experience working with a string of startups including Accretive Health, Quid, and Stride Health.(26:06) Leonard is the founding team member of Primer AI, a startup that applies state-of-the-art NLP techniques to build machines that read and write, back in early 2015.(30:31) Leonard discussed the technical challenges to develop algorithms that power Primer’s products to scale across languages other than English.(34:28) Leonard unpacked his technical post "Russian NLP” on Primer’s blog.(38:17) Leonard talked about the advances in the NLP research domain that he is most excited about in 2020 (XLNet >>> BERT).(41:10) Leonard discussed the challenges of scaling the data-driven culture across Primer AI as the company grows.(46:20) Leonard mentioned different use cases of Primer for clients in finance, government, and corporate.(51:41) Leonard talked about his decision to leave Primer and become a Data Science Health Innovation Fellow at the Berkeley Institute for Data Science.(54:30) Leonard went over applications of data science in healthcare that will be adopted widely in the next few years.(1:02:45) Leonard discussed his process of writing a book called “Data Science Bookcamp.”(1:07:21) Leonard revealed how he chose the case studies to be included in the book.(1:10:27) Closing segment.His Contact Info:LinkedInGoogle ScholarBerkeley Institute For Data ScienceHis Recommended Resources:Semi-Supervised LearningAssociation Rule LearningspaCy (Open-Source Library for Advanced NLP)fastText  (NLP library from Facebook)XLNet: Generalized Autoregressive Pretraining for Language UnderstandingBERT: Pretraining of Deep Bidirectional Transformers for Language UnderstandingFederated Learning with Differential Privacy: Algorithms and Performance AnalysisDifferential Privacy- Enabled Federated Learning for Sensitive Health DataOasis Labs and Dr. Dawn SongFitbit and Apple WatchWalter Pitts who invented neural networksPaul Werbos who invented back-propagationFei-Fei Li who constructed the ImageNet dataset“The Signal and The Noise” by Nate SilverYou can read the completed chapters of "Data Science Bookcamp" using the codes below:Permanent discount code: poddcast195 free eBook codes: dcdsprf-B373, dcdsprf-CA3B, dcdsprf-299E, dcdsprf-6E5, and dcdsprf-9660 (activated and will last for 2 months)

    Episode 28: Excelling in Data Analytics with Vincent Tatan

    Play Episode Listen Later Mar 1, 2020 58:11


    Show Notes:(2:25) Vincent talked about his educational background, in which he studied Information Systems at the Singapore Management University. (4:20) Vincent talked about the capstone project on Data Analytics Practicuum that he completed for his degree. (5:47) Vincent went over his Software Development and Business Intelligence internship experience with VISA.(9:30) Vincent landed a Data Science internship at Lazada Group, an international e-commerce company based in Singapore.(11:09) Vincent decided to come back to VISA for a full-time software engineering role after college.(12:32) Vicent worked on a variety of projects from designing micro-services to developing data dictionary management system during his 2-year stint at VISA(15:30) Vincent shared Nigel Poulton's resources on Kubernetes and Docker.(16:38) Vincent discussed his career move to become a Data Analyst at Google Singapore, focusing on Trust and Safety.(20:16) Vincent went over the unique challenges of fighting abuse at Google’s scale.(22:55) In the project "Stock Analysis with Pandas and Scikit-Learn," Vincent walked through an application that retrieves and displays the right financial insights quickly about a certain company stocks price.(25:11) In the project “Build Your Own Data Dashboard,” Vincent showed a tutorial on how to work with Dash, an open-source Python library to build web apps which are optimized for data visualization.(29:13) In the project “Deploy Your First Analytics Project,” Vincent showed a tutorial on how to deploy a dashboard web app with Heroku.(31:48) Vincent shared tips on how to ace the data analysis and data science interviews from big companies in his article “Ace Your Data Analytics Interviews.”(36:15) Vincent unpacked his article “Data Analytics Is Hard… Here’s How To Excel."(43:52) Vincent unpacked his article "How I Overcome Imposter Syndrome in Data Analytics.”(51:11) Vincent shared his thoughts regarding the tech and data community in Singapore.(53:42) Closing Segment. His Contact Info:LinkedInTwitterGitHubMediumYouTubeHis Recommended Resources:Nigel Poulton on KubernetesSebastian ThrunHans RoslingDJ Patil“Deep Work” by Cal Newport

    Episode 27: Feature Engineering with Ben Fowler

    Play Episode Listen Later Jan 24, 2020 60:50


    Show Notes:(2:17) Ben talked about his past career working in the golf industry - working at the National Golf Foundation and the PGA of America.(4:12) Ben discussed about his first exposure to machine learning and data science.(5:06) Ben talked about his motivation for pursuing an online Master’s degree in Data Science at Southern Methodist University.(6:02) Ben emphasized the importance of a Data Mining course that he took.(8:12) Ben discussed his job as a Senior Data Scientist at CarePredict, an AI elder care platform that helps senior live independently, economically, and longer.(8:56) Ben shared his thought about data security, the biggest challenge of adopting machine learning in healthcare.(10:38) Ben talked about his next employer JM Family Enterprises, one of the largest companies in the automotive industry.(12:44) Ben walked through the end-to-end model development process to solve various problems of interests in his Data Scientist work at JM Family Enterprises.(14:15) Ben discussed the challenges around feature engineering and model experiments in this process.(18:09) Ben shared information about his current role as Machine Learning Technical Lead at Southeast Toyota Finance.(19:29) Ben talked about his passion to do IC data science work.(22:37) Ben went over different conferences he has been / will be at.(26:03) Ben shared the best practices/techniques/libraries to do efficient feature engineering and feature selection, as presented at Palm Beach Data Science Meetup in September 2018 and PyData Miami in January 2019.(29:27) Ben talked about the importance of doing exploratory data analysis and logging experiments before engaging in any feature engineering / selection work.(32:50) Ben shared his experiments performing data science for Fantasy Football - specifically using machine learning to predict the future performance of players, from his talk at the Palm Beach Data Science Meetup last year.(37:25) Ben talked about his experience using H2O AutoML.(40:07) Ben gave a glimpse of his talks about evaluating traditional and novel feature selection approaches at PyData LA and Strata Data Conf.(51:25) Ben gave his advice for people who are interested in speaking at conferences.(52:29) Ben shared his thoughts about the tech and data community in the greater Miami area.(53:16) Closing Segment.His Contact Info:LinkedInHis Recommended Resources:MLflow from DatabricksStreamlit LibraryPyData ConferenceH2O World ConferenceO’Reilly Strata Data and AI ConferenceREWORK Summit ConferencePandas LibraryXGBFir Librarytsfresh LibraryLending Club DatasetSHAP library from Scott Lundberg"Interpretable Machine Learning with XGBoost" by Scott LundbergAmazon SageMakerGoogle Cloud AutoMLH2O AutoMLWes McKinney’s "Python for Data Analysis"

    Episode 26: From Cognitive Neuroscience To Reinforcement Learning with Arthur Juliani

    Play Episode Listen Later Jan 8, 2020 50:28


    Show Notes:(2:00) Arthur talked about his undergraduate studying Psychology at North Carolina State University.(3:28) Arthur mentioned his time working as a research assistant at the LACElab in NCSU that does human factor and cognition research.(5:08) Arthur discussed his decision to pursue a graduate degree in Cognitive Neuroscience at the University of Oregon right after college.(6:35) Arthur went over his Master's thesis (Navigation performance in virtual environments varies with fractal dimension of landscape) in more detail(10:30) Arthur unpacked his popular blog series called “Simple Reinforcement Learning in TensorFlow” on Medium.(12:56) Arthur recalled his decision to join Unity to work on its reinforcement learning problems.(14:31) Arthur recalled his choice to do the Ph.D. part-time while working full-time.(16:24) Arthur discussed problems with existing reinforcement learning simulation platforms and how the Unity Machine Learning Agents Toolkit addresses those.(18:30) Arthur went over the challenges of maintaining and continuously iterating the Unity ML Agents toolkit.(20:36) Arthur emphasized the benefit of training the agents with an additional curiosity-based intrinsic reward, which is inspired from a paper from UC Berkeley researchers (check out the Unity blog post).(22:33) Arthur talked about the challenges of implementing such curiosity-based techniques.(25:15) Arthur unpacked the introduction of the Obstacle Tower - a high fidelity, 3D, third person, procedurally generated environment - released in the latest version of the toolkit (read his blog post “On “solving” Montezuma’s Revenge”).(29:15) Arthur discussed the Obstacle Tower Challenge, a contest that offers researchers and developers the chance to compete to train the best-performing agents on the Obstacle Tower Environment.(32:49) Referring to his fun tutorial called “GANs explained with a classic sponge bob squarepants episode,” Arthur walked through the theory behind the Generative Adversarial Network algorithm via an explanation using an episode of Spongebob Squarepants.(34:30) Arthur extrapolated on his post “RL or Evolutionary Strategies? Nature has a solution: Both.”(38:36) Arthur shared a couple of approaches to balance the bias and variance tradeoff in reinforcement learning models, referring to his article “Making sense of the bias/variance tradeoff in Deep RL.”(41:19) Arthur talked about successor representations and their applications in deep learning, psychology, and neuroscience (read his post "The present in terms of the future: Successor representations in RL”).(42:38) Arthur reflected on the benefits of his Psychology and Neuroscience background for his research career.(44:21) Arthur shared his advice for graduate students who want to make a dent in the AI / ML research community.(45:30) Closing segment.His Contact Info:TwitterGitHubMediumLinkedInGoogle ScholarUnity BlogHis Recommended Resources:DeepMindGoogle BrainBeing and Time (by Martin Heidegger)

    Episode 25: Algorithmic Trading with Alexandr Honchar

    Play Episode Listen Later Dec 22, 2019 56:26


    Show Notes:(2:27) Alex talked about his undergraduate experience studying Applied Mathematics at the Kiev Polytechnic Institute.(3:15) Alex quickly went over his time working remotely as a Machine Learning Engineer for a US-company called Inma AI during university.(4:24) Alex mentioned his decision to pursue a Master’s degree in Mathematics at the University of Verona.(6:21) Alex went over the Math graduate classes that he took for his degree, including differential geometry and optimization theory.(7:58) Alex talked about his experience working at Mlvch to create the best solutions on the market related to visual style transfer and image enhancement.(10:01) Alex shed some light on his Master’s thesis work called “Boosting financial models calibration with deep neural networks” (Read his post "Meta-learning in finance: boosting models calibration with deep learning”).(11:39) Alex talked about the applications of meta-learning, a powerful technique to train deep neural networks, in physics and biology.(13:15) Alex shared his experience working as a partner and solutions architect at Mawi Solutions, a Ukraine-based hardware-and-software company that is disrupting the market of wearable preventive healthcare.(16:29) In reference to blog post “Deep learning: the final frontier for signal processing and time series analysis?,” Alex discussed ways to apply deep learning to model time series (Watch his talk at PyCon Italia 2019 as well).(18:52) Alex is currently the Co-Founder and CTO of Neurons Lab, an innovative European AI boutique based out of London that serves clients in financial tech, marketing tech, and medical tech areas.(22:41) Alex is well-known for a series of blog posts that experiment neural networks for algorithmic trading time series forecasting (Check out his Deep Trading code repo).(26:51) Alex emphasized the benefits of using multi-task learning, in reference to his post “Multitask learning: teach your AI more to make it better” (check out his experiments for 4 different use cases).(29:35) Alex advocated for the need of using generative models in applied AI (Read his 2 blog posts “GANs beyond generation: 7 alternative use cases” and “Generative AI: A key to machine intelligence?”).(34:02) Alex talked about a new approach called disentangled representation learning, which combines the best of classical math and machine learning modeling, in reference to his article “GANs” vs “ODEs”: the end of mathematical modeling? (See his Code and Talk).(36:36) In reference to his post “Fantastic data scientists: where to find them and how to become one,” Alex described the type of data scientist he identifies with as well as the skills that he is looking to improve upon.(42:08) Alex went over the evolution of algorithms for asset portfolio management, referring to his piece “AI for portfolio management: from Markowitz to Reinforcement Learning” (See his Code Repo).(45:09) Alex gave his advice for people who want to get into blogging and public speaking. (48:00) Alex gave his AI predictions for the year 2020 (Read his 2018 predictions for researchers and for developers).(49:54) Alex shared his thoughts regarding the data science community in East Europe vs West Europe.(52:10) Closing Segment.His Contact Info:LinkedInMediumTwitterGitHubHis Recommended Resources:Quantopian Lectures on Quantitative InvestmentSuccessful Algorithmic Trading Ebook from QuantStartAdvanced Algorithmic Trading Ebook from QuantStart"Python for Finance” by Yves Hilpisch"Derivatives Pricing in Python” by Yves Hilpisch"Advances in Financial Machine Learning” by Marcos Lopez de Prado“Pattern Recognition and Machine Learning” by Christopher BishopMachine Learning and Reinforcement Learning in Finance in CourseraOpenAIDeepMindSalesforce EinsteinFacebook AIGoogle AI

    Episode 24: From Actuarial Science to Machine Learning with Mael Fabien

    Play Episode Listen Later Dec 9, 2019 71:43


    Show Notes:(2:08) Mael recalled his experience getting a Bachelor of Science Degree in Economics from HEC Lausanne in Switzerland.(4:47) Mael discussed his experience co-founding Wanago, which is the world’s first van acquisition and conversion crowdfunding platform.(9:48) Mael talked about his decision to pursue a Master’s degree in Actuarial Science, also at HEC Lausanne.(11:51) Mael talked about his teaching assistantships experience for courses in Corporate and Public Finance.(13:30) Mael talked about his 6-month internship at Vaudoise Assurances, in which he focused on an individual non-life product pricing.(16:26) Mael gave his insights on the state of adopting new tools in the actuarial science space.(18:12) Mael briefly went over his decision to do a Post Master’s program in Big Data at Telecom Paris, which focuses on statistics, machine learning, deep learning, reinforcement learning, and programming.(20:51) Mael explained the end-to-end process of a deep learning research project for the French employment center on multi-modal emotion recognition, where his team delivered state-of-the-art models in text, sound, and video processing for sentiment analysis (check out the GitHub repo).(26:12) Mael talked about his 6-month part-time internship doing Natural Language Processing for Veamly, a productivity app for engineers.(28:58) Mael talked about his involvement with VIVADATA, a specialized AI programming school in Paris, as a machine learning instructor.(34:18) Mael discussed his current responsibilities at Anasen, a Paris-based startup backed by Y Combinator back in 2017.(38:12) Mael talked about his interest in machine learning for healthcare, and his goal to pursue a Ph.D. degree.(40:00) Mael provided a neat summary on current state of data engineering technologies, referring to his list of in-depth Data Engineering Articles.(42:36) Mael discussed his NoSQL Big Data Project, in which he built a Cassandra architecture for the GDELT database.(47:38) Mael talked about his generic process of writing technical content (check out his Machine Learning Tutorials GitHub Repo).(52:50) Mael discussed 2 machine learning projects that I personally found to be very interesting: (1) a Language Recognition App built using Markov Chains and likelihood decoding algorithms, and (2) the Data Visualization of French traffic accidents database built with D3, Python, Flask, and Altair.(56:13) Mael discussed his resources to learn deep learning (check out his Deep Learning articles on the theory of deep learning, different architectures of deep neural networks, and the applications in Natural Language Processing / Computer Vision).(57:33) Mael mentioned 2 impressive computer vision projects that he did: (1) a series of face classification algorithms using deep learning architectures, and (2) face detection algorithms using OpenCV.(59:47) Mael moved on to talk about his NLP project fsText, a few-shot learning text classification library on GitHub, using pre-trained embeddings and Siamese networks.(01:03:09) Mael went over applications of Reinforcement Learning that he is excited about (check out his recent Reinforcement Learning Articles).(01:05:14) Mael shared his advice for people who want to get into freelance technical writing.(01:06:47) Mael shared his thoughts on the tech and data community in Paris.(01:07:49) Closing segment.His Contact Info:TwitterWebsiteLinkedInGitHubMediumHis Recommended Resources:Deep Learning by Ian Goodfellow, Yoshua Bengio, and Aaron CourvillePyImageSearch by Adrian RosebrockStation F Incubator in ParisBenevolentAIEconometrics Data Science: A Predictive Modeling Approach by Francis Diebold

    Episode 23: Machine Learning for Finance with Jannes Klaas

    Play Episode Listen Later Dec 3, 2019 42:45


    Show Notes:(1:57) Jannes discussed his undergraduate experience studying International Business Administration at Rotterdam School of Management - Erasmus University (one of Europe’s top 10 research-based business schools).(3:06) Jannes talked about his work on the IHS Global Green City Index, a set of measurements that can give majors actionable insight in where cities stand in sustainability.(4:05) Jannes talked about his involvement with the Turing Society in Rotterdam, where he developed a new kind of machine learning class called “Bletchley Bootcamp for Machine Learning in Financial Context".(5:20) Jannes discussed how he built out the materials and inviting guest lectures for his machine learning for finance class.(6:57) Jannes talked about his decision to pursue a Master’s degree in Financial Economics at Said Business School, a part of Oxford University.(8:04) Jannes went over the most useful graduate course that he took for the Master’s degree, called “Information and Communication in Finance."(11:23) Jannes discussed his role with the Oxford Artificial Intelligence Society, which provides a platform to educate, build, connect, and employ an AI community that constantly drives innovation for the university and the world.(14:23) Jannes shared his thoughts regarding challenges of bringing different perspectives into the conversations around AI.(15:40) Jannes shared a brief overview of his current employer, QuantumBlack, and an example project with Formula 1 to optimize pitch stops.(18:12) Moving on to discuss his book “Machine Learning For Finance” (which introduces the study of machine learning and deep learning algorithms for financial practitioners), Jannes went over his motivation as well as the ideal audience.(19:27) Jannes went over in details the process of writing this book.(22:39) Jannes recommended a couple of resources for people who are new to Time Series forecasting.(23:57) Jannes revealed how he uses Twitter to keep up-to-date with NLP research.(25:54) Jannes explored 2 powerful financial applications of generative models: (1) perform synthetic data generation that can help with data labeling efforts and (2) generate realistic time series.(27:59) Jannes explained why private equity is an exciting playground for Reinforcement Learning models.(29:05) Jannes shared his opinions on common approaches to do Reinforcement Learning in the finance industry.(32:13) Jannes recommended the best practices to deploy machine learning models into production.(34:26) Jannes shared some interesting research projects of ethics and fairness in machine learning that attracts his attention.(36:56) Jannes shared how his financial economics background contributes to him being a good data scientist.(38:11) Jannes shared his thoughts on the tech and data community in London.(38:56) Closing segment.His Contact Info:LinkedInTwitterGitHubMediumHis Recommended Resources:Machine Learning For FinanceOxford Internet InstituteQuant GANs: Deep Generation of Financial Time Series paperMLFlow Open-Source Platform for End-to-End ML LifecycleProfessor Sandra Wachter’s paper: “Affinity Profiling and Discrimination by Association in Online Behavioral Advertising"Two SigmaBenevolentAIComputer Age Statistical Inference by Bradley Efron and Trevor Hastie

    Episode 22: Leading Self-Driving Cars Projects with Jan Zawadzki

    Play Episode Listen Later Nov 20, 2019 72:48


    Show Notes:(2:15) Jan discussed his undergraduate experience studying Business Administration and Economics from Goshen College in Indiana.(3:48) Jan went over his first job out of college: working as a Strategy and Enterprise Intelligence consultant at the EY office in Berlin, a Big 4 consulting firm.(6:09) Jan talked about his decision to pursue a part-time Master’s degree in Computer Science at the Trier University of Applied Sciences while working at EY.(7:21) Jan covered the most useful graduate courses during his Master’s degree, including Advanced Programming and Distributed Systems.(9:05) As part of his program, Jan did his thesis with the Scout24. In fact, he even wrote a blog post offering a glimpse of what it’s like to be a data scientist at Scout24.(12:31) Jan discussed the benefits of taking deep learning online classes from Andrew Ng’s deeplearning.ai platform.(15:52) Jan is also a mentor and ambassador with deeplearning.ai, in which he gives feedback on the educational content, discusses new product ideas, and writes forum entries.(19:30) Jan discussed his current projects at Carmeq GmbH, the Berlin-based innovation vehicle of Volkswagen AG.(21:22) Related to his work at Carmeq, in the blog post “The State of Self-Driving Cars for Everybody,” Jan outlined the 6 main infrastructural problems of self-driving cars for the masses. We discussed these problems in finer detail.(30:52) Jan gave a curated list of 5 mindset-chasing books that helps him become a better data scientist, referring to his article “Top 5 Business-Related Books Every Data Scientist Should Read.”(38:32) Jan shared the 5 pitfalls that young data scientists can stumble upon in their first job, referring to his article “The Power of Goal-Setting in Data Science.”(44:38) Jan emphasized the importance of using Google’s goal-setting method OKRs (Objects and Key Results) to set a data science project up for success, referring to his article “The Power of Goal-Setting in Data Science.”(48:32) Jan explained the AI Project Canvas, which answers the most pressing questions about the outcome and resources needed for an AI project.(51:50) Jan went over the importance of learning business basics for data scientists.(56:30) Referring to his post called “Becoming a Level 3.0 Data Scientist,” Jan discussed his current career trajectory as well as skills that he is looking to develop.(58:20) Jan gave his advice for data scientists to make a leap from an individual contributor to a manager.(01:02:04) Referring his post called “The Secrets to a Successful AI Strategy,” Jan gave his advice for data scientists to collaborate productively with their counterparts in product management and business operations.(01:02:55) Jan shared his opinions on the technology and data community in Berlin.(01:05:00) Closing segment.His Contact Info:MediumLinkedInTwitterGitHubHis Recommended Resources:Deep Learning Specialization from deeplearning.aiFederated LearningLex Friedman’s “Deep Learning for Self-Driving Cars” class taught at MITNassim Taleb’s “Skin In The Game"Nassim Taleb’s “Black Swan"Peter Thiel’s “Zero To One"Eric Ries’ “The Lean Startup"Daniel Kahneman’s “Thinking, Fast and Slow"Richard Rumelt’s “Good Strategy, Bad Strategy"John Doerr’s “Measure What Matters"AI Project CanvasWilliam Oncken’s “Managing Management Time"Google AI and DeepMindDeepMind: The Podcast hosted by Hannah FryHans Rosling’s “Factfulness"

    Episode 21: How To Be A Data Science Writer with Admond Lee

    Play Episode Listen Later Oct 30, 2019 79:33


    Show Notes:(2:17) Admond described his undergraduate experience studying Applied Physics from Nanyang Technological University in Singapore.(3:02) Admond talked about the importance of learning multivariate calculus and linear algebra in college.(3:38) Admond shared his research internship experience at CERN, the European Organization for Nuclear Research, in Geneva after his junior year.(4:54) Admond went over his part-time data analytics internship at SMRT Corporation while still at school.(6:10) Admond shared his decision to graduate one semester early to pursue a Data Science internship at Quantum Inventions.(7:42) Admond wrote a Medium blog post titled “My first data scientist internship” detailing his experience at Quantum Inventions.(9:07) Admond gave his advice for job seekers to filter out the signal and the noise in data science job postings, with reference to his article “Why did I reject a data scientist job?”(11:12) Admond discussed his work as a research engineer at the online gaming platform Titansoft, where he focused on Artificial Intelligence in human behavior imitation to enhance the current automation system.(13:53) Admond shared the lessons he wrote about in the blog post called “5 Lessons I Have Learned from Data Science in real working experience."(17:34) Admond talked about his current employer Micron Technology.(19:06) Admond walked through the 4 stages to define a problem statement, as evidence in his article “How to ask the right question as a data scientist.”(24:22) Admond argued for the importance of resourcefulness as a data scientist, regarding his piece “Be resourceful - one of the most important skills to succeed in data science.”(28:34) Admond went over the 5 core principles that are useful to his writing from his article “How do I write about Data Science on Medium.”(33:50) Admond considered specializing in Natural Language Processing after having been a generalist for 2 years (“Why You Should Be a Generalist first, Specialist later as a Data Scientist”).(36:11) Admond went over the speech recognition side project that he has been working on.(39:34) Admond shared his thoughts regarding ethics and privacy within speech recognition.(42:42) Admond shared other projects he’s been involved with, including public speaking and consulting gigs.(45:53) Admond talked about his involvement with the AI Time Journal as a committee member for the AI for Education 2019 Initiative, which identifies and showcases the most impactful and beneficial applications of Artificial Intelligence in the field of Education.(50:24) Admond talked about the tech and data community in Singapore.(57:38) Admond discussed his lowest point during his journey into data science and how he handled that from a psychological standpoint.(01:03:18) Admond shared the 6 lessons for recent college grads, in reference to his article “One Year After Graduation, Here is What I’ve Learned.”(01:11:24) Closing segment.His Contact Info:LinkedInMediumGitHubTwitterHis Recommended Resources:GrabBitcurate"The Lean Startup” by Eric LiesAI SingaporeTensorFlow and Deep Learning Meetup at Singapore

    Claim DataCast

    In order to claim this podcast we'll send an email to with a verification link. Simply click the link and you will be able to edit tags, request a refresh, and other features to take control of your podcast page!

    Claim Cancel