Podcast appearances and mentions of jennifer marsman

There are two common ways in which developers are incorporating proprietary and domain-specific data when building applications of Large Language Models (LLMs): Retrieval-Augmented Generation (RAG) and Fine-Tuning. RAG augments the prompt with the external data, while fine-Tuning incorporates the additional knowledge into the model itself. However, the pros and cons of both approaches are not well understood. In this paper, we propose a pipeline for fine-tuning and RAG, and present the tradeoffs of both for multiple popular LLMs, including Llama2-13B, GPT-3.5, and GPT-4. Our pipeline consists of multiple stages, including extracting information from PDFs, generating questions and answers, using them for fine-tuning, and leveraging GPT-4 for evaluating the results. We propose metrics to assess the performance of different stages of the RAG and fine-Tuning pipeline. We conduct an in-depth study on an agricultural dataset. Agriculture as an industry has not seen much penetration of AI, and we study a potentially disruptive application - what if we could provide location-specific insights to a farmer? Our results show the effectiveness of our dataset generation pipeline in capturing geographic-specific knowledge, and the quantitative and qualitative benefits of RAG and fine-tuning. We see an accuracy increase of over 6 p.p. when fine-tuning the model and this is cumulative with RAG, which increases accuracy by 5 p.p. further. In one particular experiment, we also demonstrate that the fine-tuned model leverages information from across geographies to answer specific questions, increasing answer similarity from 47% to 72%. Overall, the results point to how systems built using LLMs can be adapted to respond and incorporate knowledge across a dimension that is critical for a specific industry, paving the way for further applications of LLMs in other industrial domains. 2024: M. A. D. L. Balaguer, Vinamra Benara, Renato Luiz de Freitas Cunha, Roberto de M. Estevao Filho, Todd Hendry, Daniel Holstein, Jennifer Marsman, Nick Mecklenburg, S. Malvar, Leonardo Nunes, Rafael Padilha, Morris Sharp, B. Silva, Swati Sharma, Vijay Aski, Ranveer Chandra https://arxiv.org/pdf/2401.08406

This week: Jon Dettling from Quantis on how business can innovate to tackle supply chain challenges in the agriculture sector, including game-changing plant-based alternatives to meat. And Hilton’s Daniella Foster and Microsoft’s Jennifer Marsman outline how they engage their senior colleagues on the business case for sustainability, how to measure positive impacts, and the importance of ‘story-telling’ to keep communications relevant. Plus: fallout from the latest IPCC report; benefits of switching subsidy to renewable energy technology; and, M&S and Aldi’s latest packaging innovations to remove single use plastics, in the news digest. Hosted by Ian Welsh

Chris caught up with Jennifer Marsman, Principal Engineer on the AI for Earth team at Microsoft, right before her speech at Applied Machine Learning Days 2019 in Lausanne, Switzerland. She relayed how the team came into being, what they do, and some of the good deeds they have done for Mother Earth. They are giving away $50 million (US) in grants over five years! It was another excellent example of AI for good!

Chris caught up with Jennifer Marsman, Principal Engineer on the AI for Earth team at Microsoft, right before her speech at Applied Machine Learning Days 2019 in Lausanne, Switzerland. She relayed how the team came into being, what they do, and some of the good deeds they have done for Mother Earth. They are giving away $50 million (US) in grants over five years! It was another excellent example of AI for good!

Daniella Foster, senior director for global corporate responsibility at Hilton, and Jennifer Marsman, principal software development engineer at Microsoft, talk with Innovation Forum's Ian Welsh about how to make the business case to corporate boards so they sit up and engage with developing sustainability to future-proof business models. Investing in how to measure impacts better, both environmental and social, is crucial, they argue, alongside thinking about what the business will look like in the future.

This week: Daniella Foster, senior director for global corporate responsibility at Hilton, and Jennifer Marsman, principal software development engineer at Microsoft, talk with Ian Welsh about how to piece together the narratives for future-proofing business models. They discuss the business cases that really make sceptical board directors sit up and help drive forward smart solutions. Foster and Marsman debate how to get beyond the obvious answers and make billion-dollar savings. Plus: why California will require women directors on corporate boards; initiatives on gender empowerment, modern slavery, and food and land use at the UN General Assembly; and, changing public attitudes to plastics, in the news digest. Hosted by Ian Welsh

Hello and welcome back to Tech Forward! On this week’s episode, I spoke with Jennifer Marsman, the Principal Software Engineer of Microsoft’s “AI for Earth” group. Working with the group, she uses data science, machine learning, and artificial intelligence to aid with clean water, agriculture, biodiversity, and climate change. Jennifer has been recognized as one of the “top 100 most influential individuals in artificial intelligence and machine learning” by Onalytica, reaching the #2 slot in 2018. We discussed the work she has led with AI for Earth, the issue of bias, and her insights about building a welcoming culture in the tech sector. By running a global grant program, developing their own APIs, and funding projects, AI for Earth is, in Jennifer’s words, “a chance to use my passion for machine learning to make a difference.” One such project is FarmBeats, which is interested in sustainable, long-term solutions to maximize yield and, ultimately, reduce world hunger. To do this, they synthesize IoT, drones, machine learning, and cutting edge networking research to create precision agricultural techniques. A combination of smart sensors and aerial imagery — from drones or large helium balloons — can create a moisture map of any given field, and enable precision irrigation. Similarly, precision pesticide application benefits the environment, the farmer’s budget, and the consumer. Another initiative, Project Premonition, puts mosquitoes to work in order to predict outbreaks of diseases. Mosquitoes collect a large distribution of blood samples, and by analyzing those samples within their genomics pipeline, Project Premonition can determine the animal the blood came from and any diseases it might be carrying to generate a real-time health map of an area. Under the current model, where doctors report local cases of disease to a larger agency such as the Centers for Disease Control, it can take weeks or even a month to identify an outbreak. Using machine learning every step of the way — from placement of the traps, to making sure mosquitoes are trapped with 90% accuracy, to analyzing the data of the blood samples — Project Premonition aims to streamline and accelerate that process. While machine learning is obviously a powerful tool, there is always the danger of bias. “Any bias in the historical data that you use to make future predictions will be captured in your machine learning algorithm. You have to make sure that your users know how the data was collected. Your model is only as  good as the data you train it with.” Good intentions, such as trying to identify at-risk students and preemptively give them more support, can turn into a self-fulfilling prophecy. This is yet another strong argument in favor of diverse perspectives at all stages of product development. When it comes to tech’s diversity issues, Jennifer identified two distinct, but related issues: attracting a diverse workforce, and maintaining that diversity. She emphasized the importance of media representation — from computer games with playable female protagonists, to television shows with women in tech roles — to her own personal career journey as well as the journeys of young women today. From her internship days at Ford Motor Company, she has witnessed firsthand how crucial teamwork is: not only to coding, but also to creating a welcoming environment that everyone wants to be a part of. Jennifer, thank you so much for bringing your passion and insights to the show today. Thank you also to everyone tuning in, leaving reviews, and sharing the show with your friends and colleagues. See you next week! Connect with us Website | Twitter | Instagram | Facebook

Can Artificial Intelligence help protect the Earth? Jennifer Marsman says yes! Carl and Richard talk to Jennifer about her new role working with Microsoft Research on implementing machine learning systems to support agriculture, environmental protection, biodiversity and more. Some of the projects originate within Microsoft, but there are a number of external projects supported by the team with Azure credits and funding. AI for good!Support this podcast at — https://redcircle.com/net-rocks/donations

Can Artificial Intelligence help protect the Earth? Jennifer Marsman says yes! Carl and Richard talk to Jennifer about her new role working with Microsoft Research on implementing machine learning systems to support agriculture, environmental protection, biodiversity and more. Some of the projects originate within Microsoft, but there are a number of external projects supported by the team with Azure credits and funding. AI for good!Support this podcast at — https://redcircle.com/net-rocks/donations

Episode 18 of the Modern Agile Show features an interview with Jennifer Marsman, an expert in machine learning and artificial intelligence and a principal software development engineer at Microsoft. Jennifer spoke at YOW! 2017 about her experiment to use machine learning to analyze brain wave data to determine if people were telling the truth. She explains how supervised machine learning and deep learning via neural networks are helping make people awesome. Finally, she explains how Microsoft's culture is changing, including how they are encouraging people to conduct experiments and have a growth mindset.

We talk with Jennifer Marsman about using machine learning to build a lie detector and other cool ML projects. LinkedIn helps you write your resume. And a free chaos engineering book.

Are your machines learning? While at NDC in London, Carl and Richard hosted a panel of Jennifer Marsman, Evelina Gabasova and Barbara Fusinska talking about their work in machine learning. The conversation ranges over the variety of projects they've worked on, including using machine learning to analyze Stack Overflow data, building a machine learning-driven lie detector, doing sales data analysis with R, and more. The machine learning space is vast, and there are plenty of opportunities for everyone, you need to get exploring!Support this podcast at — https://redcircle.com/net-rocks/donations

Are your machines learning? While at NDC in London, Carl and Richard hosted a panel of Jennifer Marsman, Evelina Gabasova and Barbara Fusinska talking about their work in machine learning. The conversation ranges over the variety of projects they've worked on, including using machine learning to analyze Stack Overflow data, building a machine learning-driven lie detector, doing sales data analysis with R, and more. The machine learning space is vast, and there are plenty of opportunities for everyone, you need to get exploring!Support this podcast at — https://redcircle.com/net-rocks/donations

Jennifer Marsman is a Principal Developer Evangelist in Microsoft's Developer and Platform Evangelism group, where she educates developers on Microsoft's new technologies.  In this role, Jennifer is a frequent speaker at software development conferences around the world.  In 2016, Jennifer was recognized as one of the “top 100 most influential individuals in artificial intelligence and machine learning” by Onalytica.  She has been featured in Bloomberg for her work using EEG and machine learning to perform lie detection.  In 2009, Jennifer was chosen as "Techie whose innovation will have the biggest impact" by X-OLOGY for her work with GiveCamps, a weekend-long event where developers code for charity.  She has also received many honors from Microsoft, including the Central Region Top Contributor Award, Heartland District Top Contributor Award, DPE Community Evangelist Award, CPE Champion Award, MSUS Diversity & Inclusion Award, Gold Club, and Platinum Club.  Prior to becoming a Developer Evangelist, Jennifer was a software developer in Microsoft's Natural Interactive Services division.  In this role, she earned two patents for her work in search and data mining algorithms.  Jennifer has also held positions with Ford Motor Company, National Instruments, and Soar Technology.  Jennifer holds a Bachelor's Degree in Computer Engineering and Master's Degree in Computer Science and Engineering from the University of Michigan in Ann Arbor.  Her graduate work specialized in artificial intelligence and computational theory.  Jennifer blogs at http://blogs.msdn.com/jennifer and tweets at http://twitter.com/jennifermarsman.

Can you detect lies with machine learning? Jennifer Marsman can! Carl and Richard chatted with Jennifer while at the NDC conference in Oslo. Jennifer talked about gathering EEG data with Emotiv headsets to do lie detection by taking base line (known true and known false) questions and then applying Azure Machine Learning to classify the data. The conversation dives into the different machine learning techniques available on Azure and how certain algorithms are more effective on different data sets - it turns out EEG data works great with deep neural networking! There are lots of different opportunities in the machine learning space, time to check it out!Support this podcast at — https://redcircle.com/net-rocks/donations

Can you detect lies with machine learning? Jennifer Marsman can! Carl and Richard chatted with Jennifer while at the NDC conference in Oslo. Jennifer talked about gathering EEG data with Emotiv headsets to do lie detection by taking base line (known true and known false) questions and then applying Azure Machine Learning to classify the data. The conversation dives into the different machine learning techniques available on Azure and how certain algorithms are more effective on different data sets - it turns out EEG data works great with deep neural networking! There are lots of different opportunities in the machine learning space, time to check it out!Support this podcast at — https://redcircle.com/net-rocks/donations

Jennifer Marsman is all sorts of passionate about technology and specifically about machine learning.Get this. Jennifer made a lie detector using machine learning, and it actually works. And it was even featured on Bloomberg.I grabbed a few minutes of Jennifer's time to banter about a Microsoft hackathon we both attended. I hope you'll join us for this chat.

On this episode of Eat Sleep Code, guest Jennifer Marsman talks about Azure Machine Learning and how Microsoft is bringing AI to the masses.

Women in Technology In this show Dave and Larry sit down with their co-workers Beth Humphreys, Jennifer Marsman , Angela Binkowski and Rachel McKane to talk about the role of Women in Technology. Show Notes Jennifer's Women in Technology Series License to Drive - clearly not a John Hughes film Lego Serious Play Serious Play for Women at PDC Math Doesn't Suck by Danica McKellar (aka Winnie Cooper) Microsoft's Digigirlz Program Download / Listen to the Show http://thirstydeveloper.com/shows/TD033-WomenInTech.mp3 Blogs you should know No blogs this week! Ways to connect with your hosts: Dave Bost: Blog, Twitter Feed, Facebook Profile, GamerTag Larry Clarkin: Blog, Twitter Feed, Facebook Profile, GamerTag Brian Gorbett: Blog, Twitter Feed, Facebook Profile, GamerTag