Podcasts about learning research

Is TBL research stuck? A critical re-examination is needed to better understand what team-based learning actually does—and for whom. Read the accompanying article here: https://doi.org/10.1111/medu.70041

Candace Thille is an authority in learning science, educational technology, and AI-enabled learning environments. She is closing the two-way gap between the science of learning research and the hands-on practice of instruction to help students learn better. Timely and targeted feedback with the opportunity to apply that feedback is critical to learning, Thille says, and this is an area where AI supporting humans excels. She imagines a day in the not-too-distant future when human educators and AI-enabled assistants unite to help students learn faster and better than ever before. Learning is not a spectator sport, and AI can help us engage with learners – and educators – in new ways, Thille tells host Russ Altman on this episode of Stanford Engineering's The Future of Everything podcast. Have a question for Russ? Send it our way in writing or via voice memo, and it might be featured on an upcoming episode. Please introduce yourself, let us know where you're listening from, and share your question. You can send questions to thefutureofeverything@stanford.edu. Episode Reference Links: Stanford Profile: Candace Thille Connect With Us: Episode Transcripts >>> The Future of Everything Website Connect with Russ >>> Threads / Bluesky / Mastodon Connect with School of Engineering >>> Twitter/X / Instagram / LinkedIn / Facebook Chapters: (00:00:00) Introduction Russ Altman introduces guest Candace Thille, a professor of education at Stanford University. (00:03:16) Path into Learning Science How Candace became interested in improving how people learn. (00:03:47) The Science of Learning An overview of the field and why it's still developing. (00:04:42) Training Educators How learning science is applied in teacher education. (00:05:17) The Research to Practice Gap Why insights from classrooms rarely feed back into research. (00:06:43) Technology Supporting Teachers Using AI and other technological tools to enhance teaching. (00:09:00) The Open Learning Initiative (OLI) The origins of one of the first large-scale digital learning systems. (00:11:08) Learning with OLI How feedback and structured practice improved student outcomes. (00:13:14) Building OLI Across Disciplines The collaboration between researchers, instructors, and engineers. (00:14:36) The Accelerated Learning Study Evidence that students can learn faster without sacrificing outcomes. (00:18:02) Learning Science at Amazon Applying learning science research to workplace education. (00:22:29) Research as a Feedback Loop Why teaching practice should continuously inform research. (00:24:49) The Importance of Infrastructure Using captured learning data to improve instruction at scale. (00:25:37) Predictive AI for Learning Science The applications of older AI models in learning science research. (00:28:22) Generative AI as a Learning Interface How generative AI can make education more accessible. (00:31:01) The Myth of Learning Styles The misconception that most people have different learning styles. (00:33:30) Future In a Minute Rapid-fire Q&A: new tools, data infrastructure, and supporting learners. (00:35:24) Conclusion Connect With Us:Episode Transcripts >>> The Future of Everything WebsiteConnect with Russ >>> Threads / Bluesky / MastodonConnect with School of Engineering >>>Twitter/X / Instagram / LinkedIn / Facebook Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

Change is emotional. Even when the strategy is solid, people still feel uncertain, skeptical, or overwhelmed, especially when the vision feels huge and the path feels unclear.In this revisited episode of The Visual Lounge (originally Episode 168), Matt sits down with Jake Gittleson, who leads McKinsey's Learning Research and Innovation Lab. Jake shares why storytelling is one of the most effective tools L&D teams have for supporting change inside organisations.Instead of trying to persuade people in one big moment, Jake explains why change stories should be shared over time, through small experiments, human insights, and incremental updates that meet people where they are. He also breaks down practical ways to gather stories through interviews, outline your narrative, and use video and audio to create connection, without needing expensive gear or a polished production setup.Learning points from the episode include:00:00 - 01:21 Introduction01:21 - 02:03 Jake's background02:03 - 04:14 How Jake started using audio and video04:14 - 07:01 What does a successful change look like07:01 - 08:45 Creativity as a tip for using video at work08.45 - 11:55 Jake's role and expertise in change and innovation11:55 - 15:11 Why human connection matters in change15:11 - 18:13 Operationalizing storytelling without big budgets18:13 - 21:13 Building the right stories21:13 - 27:10 Visual approaches to telling stories27:10 - 30:21 Capturing real voices30:21 - 39:51 Speed round39:51 - 40:46 Jake's final take40:46 OutroImportant links and mentions:Connect with Jake on LinkedIn: https://www.linkedin.com/in/jake-gittleson/Check out The Learning Geeks podcast: https://www.learninggeekspod.com/Listen to Jake's first appearance on The Visual Lounge in episode 168: https://player.captivate.fm/episode/ee9c311f-7f51-4a6c-a749-c2d7090a1274

Today, we are joined by Drs. Henry Roediger and Mark McDaniel.Henry L. Roediger III is one of the world's foremost experts on human memory and learning. Currently a distinguished professor at Washington University in St. Louis, Roediger has spent his career unlocking the mysteries of how we remember—and forget—what matters most. His pioneering experiments have revealed the surprising power of retrieval practice, the perils of false memories, and the counterintuitive strategies that lead to lasting learning. He is the co-author of the bestselling book Make It Stick: The Science of Successful Learning, which has transformed classrooms and workplaces worldwide. Known both for his experimental rigor and his gift for making science practical, Roediger's insights help learners and leaders everywhere confidently apply what science now knows about how memory really works.Mark A. McDaniel is a leading authority on how people learn, age, and remember to act on their intentions. A professor of psychological and brain sciences at Washington University in St. Louis, McDaniel's research has deepened our understanding of prospective memory, cognitive aging, and the real-world factors that help—or hinder—lasting knowledge. As co-author of Make It Stick, he brings evidence-based, classroom-tested recommendations to students, teachers, and organizations alike. McDaniel is celebrated not only for his breakthrough research, but also for his ability to translate science into actionable strategies—empowering learners to overcome distractions, boost recall, and build habits that stick for a lifetime.In this conversation, we explore the fundamental building blocks of learning and memory that challenge conventional wisdom about how we acquire knowledge. The doctors reveal why much of our traditional approach to learning is counterproductive and share insights from decades of cognitive science research. Key topics include:The three components of learning: encoding, consolidation, and retrieval processesWhy learning requires memory and how they're inextricably connectedThe counterintuitive nature of effective learning strategies and why difficulty enhances retentionInsights from memory athletes and their techniques like memory palaces and visual imageryWhy memory palace techniques work for older adults despite cognitive changesHow false memories form and what this reveals about the reconstructive nature of memoryWhy retrieval practice is superior to re-reading and highlighting for long-term retentionThe testing effect and how self-assessment drives more efficient studyingWhether you're leading training programs, designing educational curricula, or seeking to optimize your own learning, this conversation provides insights on how learning really works and offers strategies for more effective knowledge acquisition.Dr. Roediger and Dr. McDaniel's Book https://www.amazon.com/Make-Stick-Science-Successful-Learning/dp/0674729013 -Website and live online programs: http://ims-online.com Blog: https://blog.ims-online.com/ Podcast: https://ims-online.com/podcasts/ LinkedIn: https://www.linkedin.com/in/charlesgood/ Twitter: https://twitter.com/charlesgood99 Chapters:(00:00) Introduction(01:00) Tool: Personal Journeys into Memory and Learning Research(04:00) Technique: The Three Components of Learning - Encoding, Consolidation, Retrieval(07:00) Tip: Why Learning Requires Memory and Connecting to Prior Knowledge(11:00) Tool: Counterintuitive Learning Strategies and Desirable Difficulties(14:00) Technique: Memory Athletes - Techniques, Abilities, and Limitations(19:00) Tip: Memory Palaces for Older Adults and Practical Applications(24:00) Tool: Working Memory, Attention, and Cognitive Overload Management(29:00) Technique: False Memories and the Reconstructive Nature of Memory(33:00) Tip: Retrieval Practice vs Re-reading - The Testing Effect(37:00) Tool: Self-Assessment and Efficient Study Targeting(39:54) Conclusion

In this special live episode, with guests Lisa Christensen, Hillary Miller and Christopher Lind, we explore their experience and deep expertise on the topics of L&D strategy and team structures - and it’s a masterclass of a conversation. Register for L&D Next 3rd - 6th March for free today at https://360learning.com/l-and-d-next/2025/ KEY TAKEAWAYS Understand what L&D is there to achieve. Team structures have to evolve and be flexible. L&D structure has to be right for achieving L&D aims, so it may be different from other teams in the organisation. Build adaptable teams by focusing on skillsets. Leverage data and hone in on one KPI. Understand which relationships and functions you need to focus on. That will evolve. Work holistically with other areas of the business. BEST MOMENTS “I've never seen a truly centralized model, ever.” “We can get hung up on the hierarchy of things and miss out.” “You gotta know who your players are, their bench strengths.” “We need better data capabilities in learning.” “Figure out what they care about and then lean on that.” Lisa Christensen Lisa leads McKinsey & Company's Learning Design and Development Center of Excellence, a global team of design experts, designing and building the learning that develops McKinsey Partners and professionals, enabling them to deliver incredible client impact. Lisa founded and leads McKinsey's Learning Research and Innovation Lab and sits on the global learning leadership team. https://www.linkedin.com/in/lisachristensen Christopher Lind Christopher Lind is a dynamic leader at the intersection of business, technology, and human experience, serving as an executive advisor, AI ethicist and sought-after voice in the L&D space. As a former Chief Learning Officer for ChenMed and GE Healthcare, Christopher has led transformative learning strategies that enhance workforce capability and business performance. A prominent commentator, speaker, and thought leader, he is known for his forward-thinking approach to digital learning, AI, and the evolving role of technology in talent development. https://www.linkedin.com/in/christopherlind Future Focused: https://christopherlind.substack.com Hillary Miller Hillary Miller is a seasoned Learning & Development leader currently heading L&D at HCA Healthcare. With a passion for driving workforce capability and business impact, she brings extensive experience in healthcare education and leadership development. Previously, as Chief Learning Officer at Penn State Health, Hillary led enterprise-wide learning strategies, fostering a culture of continuous development and innovation. https://www.linkedin.com/in/hillarybmiller VALUABLE RESOURCES https://podcasts.apple.com/gb/podcast/the-learning-development-podcast/id1466927523 L&D Master Class Series: https://360learning.com/blog/l-and-d-masterclass-home THE HOST David James David has been a People Development professional for more than 20 years, most notably as Director of Talent, Learning & OD for The Walt Disney Company across Europe, the Middle East & Africa. As well as being the Chief Learning Officer at 360Learning, David is a prominent writer and speaker on topics around modern and digital L&D. CONTACT METHOD Twitter:  https://twitter.com/davidinlearning LinkedIn: https://www.linkedin.com/in/davidjameslinkedin L&D Collective: https://360learning.com/the-l-and-d-collective Blog: https://360learning.com/blog L&D Master Class Series: https://360learning.com/blog/l-and-d-masterclass-home

BIO: James “Jimmy” Milliron is Co-Founder & President of National Brokerage Atlantic, specializing in Wealth Enhancement, Estate Planning, and Asset Protection.STORY: Jimmy wanted to invest $100,000 in Bitcoin, but when he couldn't find an easy way to do it, he bought a car instead.LEARNING: Research and learn all you can about investment opportunities before investing. “Don't be afraid to pick up the phone and make a few calls. There's nothing like picking up the phone and talking to a real person on the other end instead of just texting them.”Jimmy Milliron Guest profileJames “Jimmy” Milliron is Co-Founder & President of National Brokerage Atlantic, specializing in Wealth Enhancement, Estate Planning, and Asset Protection. An insurance veteran, he previously served as Executive Vice President at NexTier Bank, building a $400 million premium finance portfolio. He holds a BA from VMI and various securities and insurance licenses.Worst investment everJimmy's worst investment is a mix between marrying a second wife and buying a car in 2016. He invested many resources in his second marriage, but it did not last that long.When Jimmy married his second ex-wife, he wanted to invest about $100,000 in Bitcoin. But he was busy and did not have time to research and learn more about Bitcoin. When Jimmy could not find an easy way to do it, he purchased a car instead with that cash.Lessons learnedGo the extra mile in research and learning about investment opportunities before investing.Consider all the investment options available.Actionable adviceIf you're young, seek advice from a mentor or your parents about what they would do instead of arbitrarily investing in a make-me-feel-good investment. Their guidance can be invaluable in navigating the complex world of investments.Jimmy's recommendationsJimmy recommends reading Donald Trump's Art of the Deal as a valuable resource for negotiation and decision-making.No.1 goal for the next 12 monthsJimmy's number one goal for the next 12 months is losing weight.Parting words “Thank you very much. Andrew and I wish everyone well.”Jimmy Milliron [spp-transcript] Connect with Jimmy MillironLinkedInWebsiteAndrew's booksHow to Start Building Your Wealth Investing in the Stock MarketMy Worst Investment Ever9 Valuation Mistakes and How to Avoid ThemTransform Your Business with Dr.Deming's 14 PointsAndrew's online programsValuation Master Class

How can technology reshape education to cater to every student's unique learning journey? Join us as we explore this question with Dr. Kristen DiCerbo, the Chief Learning Officer at Khan Academy, who shares her transformative insights into personalized and mastery-based learning. Discover how Khan Academy's new AI tool, "Khanmigo" is revolutionizing learning as it helps educators address the diverse needs of students worldwide, breaking free from the constraints of traditional educational models.Dr. DiCerbo highlights the challenges teachers face, such as managing large classrooms and addressing special education shortages, and how technology can alleviate these burdens. By leveraging platforms like Khan Academy, educators can provide tailored learning experiences that empower students and nurture creativity. We delve into the ways technology can reduce teacher burnout, foster stronger student-teacher connections, and transform teaching roles to be more focused on creating meaningful educational experiences and maintaining a healthy work-life balance.More About Our GuestDr. Kristen DiCerbo is the Chief Learning Officer at Khan Academy, where she leads the content, design, product management, and community support teams. Dr. DiCerbo's career has focused on embedding insights from education research into digital learning experiences. Prior to her role at Khan Academy, she was Vice-President of Learning Research and Design at Pearson, served as a research scientist supporting the Cisco Networking Academies, and worked as a school psychologist. Kristen has a Ph.D. in Educational Psychology from Arizona State University.Connect with Khan Academykhanmigo.aikhanacademy.orgGot a story to share or question you want us to answer? Send us a message!About the podcast:The KindlED Podcast explores the science of nurturing children's potential and creating empowering learning environments.Powered by Prenda Microschools, each episode offers actionable insights to help you ignite your child's love of learning. We'll dive into evidence-based tools and techniques that kindle young learners' curiosity, motivation, and well-being. Got a burning question?We're all ears! If you have a question or topic you'd love our hosts to tackle, please send it to podcast@prenda.com. Let's dive into the conversation together!Important links:• Connect with us on social • Subscribe to The Sunday Spark• Get our free literacy curriculum Interested in starting a microschool?Prenda provides all the tools and support you need to start and run an amazing microschool. Create a free Prenda World account to start designing your future microschool today. More info at ➡️ Prenda.com or if you're ready to get going ➡️ Start My Microschool

L&D leaders in large organizations often cite a lack of time as the main barrier to upskilling and reskilling. While increasing learning hours is important, the CLO Lift team found that time is not the most critical factor. Instead, the quality of learning has equal or greater influence on performance than the quantity of time spent. Rather than focusing solely on hours, business leaders and L&D professionals should emphasize value. This episode explores these insights with Lisa Christensen and Huw Newton-Hill. KEY TAKEAWAYS CLO Lift is a community of 20 CLOs who have got together to solve the biggest unresolved challenges in the L&D industry. You have to create true value before enough time will be allocated to employee learning. If you are not 100% in step with the business the training that you provide will be out of date and not relevant. The learning you deliver must genuinely enable people to make progress in their careers. L&D needs to stop only being order takers. Work more closely with managers to understand business challenges and be actively involved in finding solutions. Certain elements of people´s roles impact the way they behave. If a learning intervention cannot change that element speak up and ask those who can solve that underlying issue to do so. Create a learning culture, e.g. include skill acquisition in personal reviews. Democratize access to learning. Use generative AI to make the training more relevant in every role in every geography. Measure progress against actual business objectives. Start now. Create some small experiments and generate wins. You only need one win to create credibility and act as a springboard. BEST MOMENTS “Ensure that the whole body of L&D is in lockstep with the business.” “The more courage you lean into, the more credibility you build.” “Be bold and go forth.” “Any good action today is better than perfect action a month from now.”   CLO Lift, From Time to Value report: https://learningforum.substack.com/p/clo-lift-from-time-to-value?r=ivpaq&utm_campaign=post&utm_medium=web&triedRedirect=true   Lisa Christensen Bio: As Director of Learning Design and Innovation at McKinsey & Company, Lisa leads a global team focused on cutting-edge learning solutions. She founded McKinsey's Learning Research and Innovation Lab and is a recognized thought leader in learning. Lisa is a founding member of CLO Lift and was previously a senior leader at a learning design firm.   Huw Newton-Hill Bio: Huw leads Attensi's US office, delivering AI-powered training solutions. A former strategy consultant at BCG, he now drives growth and innovation in L&D, contributing to forums like CLO Lift. VALUABLE RESOURCES The Learning And Development Podcast - https://podcasts.apple.com/gb/podcast/the-learning-development-podcast/id1466927523 L&D Master Class Series: https://360learning.com/blog/l-and-d-masterclass-home/ ABOUT THE HOST David James  David has been a People Development professional for more than 20 years, most notably as Director of Talent, Learning & OD for The Walt Disney Company across Europe, the Middle East & Africa.  As well as being the Chief Learning Officer at 360Learning, David is a prominent writer and speaker on topics around modern and digital L&D.  CONTACT METHOD  Twitter:  https://twitter.com/davidinlearning/ LinkedIn: https://www.linkedin.com/in/davidjameslinkedin/ L&D Collective: https://360learning.com/the-l-and-d-collective/ Blog: https://360learning.com/blog/ L&D Master Class Series: https://360learning.com/blog/l-and-d-masterclass-home/  

Get to know the person behind the research in this thoughtful conversation with acclaimed education researcher John Hattie. Join host Peter DeWitt and cohost Mike Nelson as they chat with John Hattie about his journey as a lifelong learner, how his Visible Learning research has resonated globally, and his goal for every educator to have a theory of learning. And, learn about John Hattie's highlight of his week--the days he spends with his grandson. Tune into this season seven finale of the Leaders Coaching Leaders Podcast for insights from an award-winning expert on enhancing student learning and achievement.Let us know what you think!

On this episode of the Getting Smart Podcast Mason is joined by repeat guest Aaron Schorn of Unrulr and two new guests for the Getting Smart Podcast, Kwaku Aning, Director of the Center of Innovation and Entrepreneurial Thinking at the San Diego Jewish Academy and Mike Yates, The Reinvention Lab at Teach for America.  Links: Kwaku Aning LinkedIn Aaron Schorn LinkedIn Mike Yates LinkedIn San Diego Jewish Academy Reinvention Lab at Teach for America Designing Your Life MSCHF Hallcraft School Studio Unrulr Shoelace Learning Julia Dexter Changing the Subject d.school Scott Center for Entrepreneurship Bill Summers - Cañon City High School  Capstone Consortium Big Picture Learning EyeCandy Previous episode with Aaron Schorn Lauryn Hill - Sister Act 2 Performance  

Today on the show, Boston Children's Museum President and CEO Carole Charnow interviews Dr. Mitchel Resnick in the next installment of our Creativity Series. Mitch Resnick is the LEGO Papert Professor of Learning Research at the MIT Media Lab and develops new technologies and activities to engage people (particularly children) in creative learning experiences. His Lifelong Kindergarten research group developed the Scratch programming software and online community, used by millions of young people around the world.  Carole talks with Mitch about his new project OctoStudio, the value of Kindergarten style learning, the 4 P's of Creative Learning, the relationship between technology and creative learning and more.

In the context of a cost-of-living crisis and increased child poverty, this podcast hears about the growing use of food banks, how they operate and the impact this has on children whose families use them. Full show notes and links: https://www.ucl.ac.uk/ioe/news/2023/dec/what-impact-do-food-banks-schools-have-childrens-learning-rftrw-s21e02

Jason Yosinski was a founding member of Uber AI Labs. He is also a co-founder of WinscapeAI a company dedicated to using custom sensor networks and machine learning to increase the efficiency and sustainability of wind farms. Jason holds a PhD in computer science from Cornell University. We talked about his experience at Uber AI, his research in deep learning, and ML for wind farms. Subscribe to Daliana's newsletter on www.dalianaliu.com for more on data science and career. Jason's Website: https://yosinski.com/ Jason's LinkedIn: https://www.linkedin.com/in/jasonyosinski/ Daliana's Twitter: https://twitter.com/DalianaLiu Daliana's LinkedIn: https://www.linkedin.com/in/dalianaliu (00:00:00) Introduction (00:06:06) His advice for Uber ML teams (00:16:03) From research to industry (00:20:24) ML for wind farms (00:25:40) Metrics for wind energy prediction (00:29:23) Start with a small dataset (00:32:00) ML in academia vs. the industry (00:33:24) Do you need a PhD for ML? (00:38:14) Daliana's story about grad school (00:41:37) The value of a PhD (00:43:13) ML Collective (00:48:36) Technical communication (00:57:21) ML Skillsets (00:59:45) Future of machine learning (01:05:23) Personal development: Hoffman process (01:15:13) Do things that excites you

Speaker: Andrew Burns, Vice President at BTS Andrew Burns is a Vice President at BTS responsible for partnering with clients to accelerate strategy alignment and culture change initiatives. His work spans the employee lifecycle, from developing assessment centers for hiring and leader development to defining what great leadership looks like within an organization's unique context. Andrew has extensive experience working in industries ranging from Aerospace to Oil and Gas, Utilities, and Software. Host: Rachel Cooke, Brandon Hall Group™ Rachel Cooke is Brandon Hall Group's Chief Operating Officer and Principal HCM Analyst. She is responsible for business operations, including client and member advisory services, marketing design, annual awards programs, conferences and the company's project management functions. She also leads Advancing Women in the Workplace and Diversity, Equity and Inclusion initiatives, research and events. Rachel worked in the HCM research industry for 20+ years and held several key management and executive positions within the Talent and Learning Research, and Performance Improvement industries.

Today my guest is Dr. Margaret McKeown. We'll start our conversation discussing vocabulary development and explicit vocabulary instruction, including how to choose words for instruction, how to teach words so students understand them deeply, and how to help students build connections between words. Our conversation then shifts to the Questioning the Author instructional intervention, which focuses on developing comprehension through conversation and can be used to foster independence and discussion amongst students. Later, I'm joined by my colleagues Gina Dignon and Rosie Maurantonio for a conversation about how we'll bring what we learned to the classroom. ****Read a full transcript of this episode and learn more about the show and Jennifer Serravallo at JenniferSerravallo.comBringing Words to Life: Robust Vocabulary Instruction ****More about Dr. Margaret McKeown:Margaret G. McKeown, PhD, is Clinical Professor Emerita of Education at the University of Pittsburgh. Before her retirement, she was also a Senior Scientist at the University's Learning Research and Development Center. Her work addresses practical, current problems that classroom teachers and their students face. She has conducted research in the areas of learning, instruction, and teacher professional development in reading comprehension and vocabulary. Dr. McKeown is a recipient of the Outstanding Dissertation Award from the International Literacy Association, is a Fellow of the American Educational Research Association, and was inducted into the Reading Hall of Fame. She is coauthor of books including Bringing Words to Life, Second Edition; Creating Robust Vocabulary; Robust Comprehension Instruction with Questioning the Author; and Vocabulary Assessment to Support Instruction.Special thanks to Alex Van Rose for audio editing this episode. Support this showSupport the show

Advitya Gemawat ( https://www.microsoft.com/en-us/research/people/agemawat/ ) is a Machine Learning Research Engineer who currently works in the Responsible AI (RAI) team as part of the Azure Machine Learning (AML) and Azure AI Studio product lines, in the Azure AI Platform org. Prior to the RAI team, he was part of the 4th cohort of the Microsoft AI Development Acceleration Program (MAIDAP), where he worked with Azure Quality, Microsoft's Gray Systems Lab (GSL), Azure Edge & Platform, and Azure Machine Learning (AML). His research experience is at the intersection of Machine Learning & Scalable Systems. His work with GSL was identified as the Microsoft Global Hackathon Executive Challenge 2022 Winner and recipient of the Best Demonstration Award at VLDB 2022. His work with the AML team on expanding the RAI Dashboard to support Object Detection models was released in Public Preview at Microsoft Build 2023. Prior to Microsoft, Advitya graduated from UC San Diego with a Data Science major, where he contributed to Project Cerebro (a Layered Data Platform for scalable Deep Learning) and was advised by Professor Arun Kumar. There, his research won the 2021 ACM SIGMOD Student Research Abstract Competition and the Best Project Award as part of UCSD's Halıcıoğlu Data Science Institute (HDSI) Undergraduate Scholarship Program HDSI Scholarship Program. As part of an internship with VMware, his work on Massively Parallel Automated Model Building for Deep Learning was included in Apache MADlib 1.18.0 release. Opinions expressed in the video solely reflect Advitya's views and not the views of his employer. Follow Advitya on the handles below: LinkedIn: linkedin.com/in/agemawat YouTube: youtube.com/@AdvityaGemawat Microsoft Research: microsoft.com/research/people/agemawat Support the show

Peter's Film Course - https://thinkinginenglish.as.me/peters-film-course Thinking in English Classes Website - https://thinkinginenglish.link/ --------- What insights can science provide about language learning? And how can we use this research to learn English? Keep listening to find out! TRANSCRIPT - https://thinkinginenglish.blog/2023/07/19/252-science-and-english-learning-research-backed-tips-and-tricks-to-help-you-study/ ----------- My Links ⁠7 Day FREE CONVERSATION CLUB TRIAL - https://www.patreon.com/thinkinginenglish ⁠⁠ ⁠JOIN THE CONVERSATION CLUB  -- https://www.patreon.com/thinkinginenglish ⁠ ⁠Take a Class (Use code TRIAL50 for 50% off) - https://thinkinginenglish.link/⁠ ⁠Buy Me a Coffee - https://www.buymeacoffee.com/dashboard⁠ NEW YOUTUBE Channel!!! - https://www.youtube.com/@thinkinginenglishpodcast  INSTAGRAM - thinkinginenglishpodcast (https://www.instagram.com/thinkinginenglishpodcast/)   Blog - thinkinginenglish.blog -------- Vocabulary To distribute (v) - to spread out tasks, resources, or information over intervals of time, space, or people. Optimal (adj) - the most favourable or best possible. Retention (n) - the ability to retain or remember information. Recall (n) - the ability to remember things. Input (n) - information or data that is fed into a system or the brain. Memorisation (n) - the act or process of learning something so that you will remember it exactly: Literacy (n) - the ability to read and write. Immediate feedback (n) - prompt information given in response to an action or performance. --- Send in a voice message: https://podcasters.spotify.com/pod/show/thinking-english/message Support this podcast: https://podcasters.spotify.com/pod/show/thinking-english/support

Omar Flores (@OmarUFlorez en Twitter) es un experto en Modelos de Lenguaje, empleando señales sociales para optimizar la interpretación y expresión semántica de estos modelos. Sus métodos han fortalecido notablemente tareas como la búsqueda y recomendación. Ha trabajado como Machine Learning Research Scientist en Twitter, Senior Research Manager en Capital One y Research Scientist en Intel. Recibió el Premio a la Innovación de IBM Research en Análisis de Datos Escalables que financió su tesis doctoral. Este episodio profundiza en los avances y retos de la inteligencia artificial (IA), desde el modelo maestro, hasta los progresos en los modelos de lenguaje y la gestión de memoria. Se discute sobre la integración social de la IA y la capacidad de los modelos generativos para resolver problemas. Se aborda sobre memoria in context, la estrategia de IA de Perú, la diversidad de datos y las implicaciones éticas.

feedback @ ryan@soulsearching.in EPISODE LINKS: Website : https://sebastianraschka.com/ Linkedin : https://www.linkedin.com/in/sebastianraschkaTwitter : https://twitter.com/rasbtGitHub : https://github.com/rasbtYouTube : https://www.youtube.com/c/sebastianraschka PODCAST INFO: Podcast website: https://anchor.fm/ryandsouza Apple Podcasts: https://apple.co/3NQhg6S Spotify: https://spoti.fi/3qJ3tWJ Amazon Music: https://amzn.to/3P66j2B Google Podcasts: https://bit.ly/3am7rQc Gaana: https://bit.ly/3ANS4v1 RSS: https://anchor.fm/s/609210d4/podcast/rss

Many school librarians are exploring how they can use ChatGPT and generative AI to support teaching, learning, research, and creativity.  We invited three librarians to share their early discoveries and some innovative ways they are using this rapidly expanding technology. Follow on Twitter: @joycevalenza @aelissmalespina@lucasjmaxwell @sgthomas1973 @bamradionetwork @jonHarper70bd Joyce Valenza is Associate Professor, at Rutgers University, SC&I and wrote the NeverendingSearch Blog for School Library Journal (now on hiatus), and contributes to several other library and tech publications. She speaks globally about the thoughtful use of technology in learning and the power of librarians to lead. Joyce was honored with the American Association of School Librarians' Distinguished Service Award and named an AASL Social Media Leadership Luminary. She is a Milken Educator and an American Memory Fellow. Joyce earned her doctorate in information science from the University of North Texas. Lucas Maxwell has been working with youth in libraries for fifteen years. He currently works as a school librarian in South London, UK. In 2017 he was named the UK's School Librarian of the Year, and in 2022, he was called the UK's Reading for Pleasure Champion by the UK's Literacy Association. Elissa Malespina is the librarian at Union High School in Union, NJ, and serves as a Board of Education Member for the South Orange - Maplewood School District, where she has been instrumental in helping to pass policies that strengthen the district's commitment to have diverse resources and materials available to all students. Elissa is at the forefront of using Web 2.0 resources and tools like Augmented Reality to make the experience in my library more interactive. Her work has been featured in NPR, School Library Journal, Publishers Weekly, and the PBS Documentary Film School Sleuth: The Case of The Wired Classroom. *The views expressed by Malespina do not represent her employer.

Are you interested in radical collaboration as an education form? Our summary today works with the chapter titled Radical collaboration: flipping the paradigm on learning from 2021 by Kelly Boucher, from the book titled Collaboration, visionaries share a new way of living. This episode is a great preparation for our next interviewee, Kelly Boucher in episode 114. Since we are investigating the future of cities, I thought it would be interesting to see how to think otherwise of life and place. This chapter presents radically collaborative ways of living well with the world and others in glorious multiplicity and complexity. As the most important things, I would like to highlight 3 aspects: Pedagogy is a term usually used to describe the art of teaching and learning or the agency that joins teaching and learning, but it can evolve to an active dialogue and being in relation with the world. Being in relation can mean, according to Aboriginals, that the boundaries between humans and nature are blurred because everything is animate. Through being in relation and radical re-learning, we take up our responsibility and step into radical collaboration with the world. You can find the book through this link. Connecting episodes you might be interested in: No.099 - Interview with Noel Tighe about leaving ego outside for the best outcomes in design processes; No.114 - Interview with Kelly Boucher about radical collaboration and unlearning; You can find the transcript through this link. What wast the most interesting part for you? What questions did arise for you? Let me know on Twitter @WTF4Cities or on the wtf4cities.com website where the shownotes are also available. I hope this was an interesting episode for you and thanks for tuning in. Music by Lesfm from Pixabay

Welcome to The Nonlinear Library, where we use Text-to-Speech software to convert the best writing from the Rationalist and EA communities into audio. This is: Touch reality as soon as possible (when doing machine learning research), published by Lawrence Chan on January 3, 2023 on The AI Alignment Forum. Related to: Making Beliefs Pay Rent, The Feeling of Idea Scarcity, Micro-Feedback Loops and Learning, The Three Stages of Rigor, Research as a Stochastic Decision Process, Chapter 22 of HPMOR. TL;DR: I think new machine learning researchers often make one of two kinds of mistakes: not making enough contact with reality, and being too reluctant to form gears-level models of ML phenomena. Stereotypically, LW/AF researchers tend to make the former mistake, while academic and industry researchers tend to make the latter kind. In this post, I discuss what I mean by “touching reality” and why it's important, speculate a bit on why people don't do this, and then give concrete suggestions. Epistemic status: I'm pretty frustrated with how slow I write, so this is an experiment in writing fast as opposed to carefully. That being said, this is ~the prevailing wisdom amongst many ML practitioners and academics, and similar ideas have been previously discussed in the LessWrong/Alignment Forum communities, so I'm pretty confident that it's directionally correct. I also believe (less confidently) that this is good advice for most kinds of research or maybe even for life in general. Acknowledgments: Thanks to Adrià Garriga-Alonso for feedback on a draft of this post and Justis Mills for copyediting help. Introduction: Broadly speaking, I think new researchers in machine learning tend to make two kinds of mistakes: Not making contact with reality. This is the failure mode where a new researcher reads a few papers that their friends are excited about, forms an ambitious hypothesis about how to solve a big problem in machine learning, and then spends months drafting a detailed plan. Unfortunately, after months of effort, our new researcher realizes that the components they were planning to use do not work nearly as well as expected, and as a result they've wasted months of effort on a project that wasn't going to succeed. Not being willing to make gears-level models. This is the failure mode where a new researcher decides to become agnostic to why anything happens, and believes empirical results and only empirical results even when said results don't “make sense” on reflection. The issue here is that they tend to be stuck implementing an inefficient variant of grad student descent, only able to make small amounts of incremental progress via approximate blind search, and end up doing whatever is popular at the moment. That's not to say that these mistakes are mutually exclusive: embarrassingly, I think I've managed to fail in both ways in the past. That being said, this post is about the first failure mode, which I think is far more common in our community than the second. (Though I might write about the second if there's enough interest!) Here, by “touching reality”, I mean running experiments where you check that your beliefs are right, either via writing code and running empirical ML experiments, or (less commonly) grounding your ideas either in a detailed formalism (to the level where you can write proofs for new, non-trivial theorems about said ideas). I don't think writing code or inventing a formalism qualify by themselves (though they are helpful); touching reality requires receiving actual concrete feedback on your ideas. Why touch reality? I think there's four main reasons why you should do this: Your ideas may be bad When you're new to a field, it's probably the case that you don't fully understand all of the key results and concepts in the field. As a result, it's very likely the case that the ideas you come up with are bad. This is especially true for fields like machine learning that have significant amounts of tacit knowledge. ...

So it's been a couple of weeks since I've released a podcast episode, but in case you didn't know, all of the episodes are streamed on Crowdcast, and I also post them to YouTube. So the last few episodes were visual; last week's episode was based on icebreakers using slides, and before that, I streamed live from DevLearn in Las Vegas. I can't format those appropriately for a podcast, but you can find them on www.TheTLDC.com or on our YouTube channel. This episode was great for a podcast. Jo Cook did a TLDC takeover to announce her and Jane Daly releasing a research report on Hybrid and Virtual Learning. Jo is an amazing Virtual Learning facilitator, and the report findings led to lots of community discussion, which you'll hear Jo share. Give it a listen - there's lots to learn here for everyone from VILT Producers to Zoom attendees. You can download the report here: https://virtualresearchinsights.com/report2022/ And check out their infographic for the report here: https://virtualresearchinsights.com/2022/09/26/infographic-for-improving-virtual-and-hybrid-learning/

A new diversity toolkit is providing resources for long-term care organizations to help raise awareness of issues of equity, diversity and inclusion. The Embracing Diversity: A Toolkit for Supporting Inclusion in Long-Term Care Homes is an interactive resource with print and online components that gives long-term care homes practical steps to nurture diverse and welcoming communities.We spoke with Michelle Fleming and Ashley Flanagan about the toolkit and the importance of understanding issues of equity, diversity and inclusion in long-term care. Michelle Fleming is a senior knowledge broker at the Ontario Centres for Learning Research and Innovation in Long-Term Care.  Dr. Ashley Flanagan is a research fellow in Diversity in Aging at the National Institute on Aging.For more information about the toolkit, contact Michelle at: MFleming@bruyere.orgLearn more about the Strengthening a Palliative Approach to Long-Term Care project at: https://spaltc.ca/

Jak wygląda praca Machine Learning Research Engineera w Allegro? Jak udaje nam się efektywnie operować na ogromnej ilości danych oraz milionach parametrów? Czym są modele językowe trenowane na korpusie danych pochodzących z Allegro i jaki mają wpływ na realizowane projekty?Jak współpracujemy ze środowiskiem naukowym? I wreszcie - gdzie i w jakiej formie można znaleźć wiedzę, którą dzieli się nasz zespół Machine Learning Research?O tym wszystkim rozmawialiśmy z kolejnym gościem Allegro Tech Podcast, czyli Riccardo Belluzzo, który pracuje jako Machine Learning Research Engineer w Allegro.  CONTACT LINKS / LINKS MENTIONED IN THE PODCAST:https://www.linkedin.com/in/riccardo-belluzzo/Allegro ML Research - https://ml.allegro.tech/Do you speak Allegro?” Large Scale Language Modeling - https://www.youtube.com/watch?v=6T-R4kgIbBsHerBERT na GitHub - https://github.com/allegro/HerBERT BIO:Riccardo is a Research Engineer in the Allegro ML team and specializes in Natural Language Processing and Understanding. Riccardo is also a music freak, playing guitar in his free time and running a podcast about underground music and emerging artists.

In this episode, we speak to Shawna Young and Mitchel Resnick of the Scratch Foundation, which runs the largest creative computing community in the world around the Scratch programming language. Recommended Resources:Mitch Resnick and Ken Robinson, Lifelong Kindergarten2021 Scratch Foundation Annual ReportShawna Young is the Executive Director of the Scratch Foundation. Before coming to Scratch, Young led the Duke University Talent Identification Program (Duke TIP), one of the largest academic talent searches, with over 450,000 K-12 students and over 3 million alumni. She also spearheaded the expansion of the Office of Engineering Outreach Programs (OEOP) at MIT, serving as the Executive Director for eight years.  The OEOP provides rigorous educational opportunities in science, technology, engineering, and mathematics (STEM) to K-12 students from primarily underrepresented and underserved backgrounds. Young started her career as a public high school science teacher in North Carolina, then working as a curriculum developer at the Educational Development Center.Mitchel Resnick is the LEGO Papert Professor of Learning Research and Director of the Lifelong Kindergarten group at the Massachusetts Institute of Technology (MIT) Media Lab, which developed the Scratch programming software and online community, the world's leading coding platform for kids. His group has also collaborated for many years with the LEGO Company and the LEGO Foundation on the development of new educational ideas and products, including LEGO Mindstorms and LEGO WeDo robotics kits. Resnick co-founded the Computer Clubhouse project, an international network of 100 after-school learning centers, where youth from low-income communities learn to express themselves creatively with new technologies.

This week, Anna (https://twitter.com/annarrose) and Tarun (https://twitter.com/tarunchitra) chat with Florian Tramèr (https://twitter.com/florian_tramer), Assistant Professor at ETH Zurich (https://ethz.ch/en.html). They discuss his earlier work on side channel attacks on privacy blockchains, as well as his academic focus on Machine Learning (ML) and adversarial research. They define some key ML terms, tease out some of the nuances of ML training and models, chat zkML and other privacy environments where ML can be trained, and look at why the security around ML will be important as these models become increasingly used in production. Here are some additional links for this episode: * Episode 228: Catch-up at DevConnect AMS with Tarun, Guillermo and Brendan (https://zeroknowledge.fm/228a/) * Florian Tramèr's Github (https://github.com/ftramer) * Florian Tramèr's Publications & Papers (https://floriantramer.com/publications/) * ETH Zurich (https://ethz.ch/en.html) * DevConnect (https://devconnect.org/) * Tarun Chritra's Github (https://github.com/pluriholonomic) * Single Secret Leader Election by Dan Boneh, Saba Eskandarian, Lucjan Hanzlik, and Nicola Greco (https://eprint.iacr.org/2020/025) * GasToken: A Journey Through Blockchain Resource Arbitrage by Tramèr, Daian, Breidenbach and Juels (https://floriantramer.com/docs/slides/CESC18gastoken.pdf) * Enter the Hydra: Towards Principled Bug Bounties and Exploit-Resistant Smart Contracts by Tramèr, Daian, Breidenbach and Juels (https://eprint.iacr.org/2017/1090) * Ronin Bridge Hack – Community Alert: Ronin Validators Compromised (https://roninblockchain.substack.com/p/community-alert-ronin-validators?s=w) * InstaHide: Instance-hiding Schemes for Private Distributed Learning, Huang et al. 2020. (https://arxiv.org/abs/2010.02772) * Is Private Learning Possible with Instance Encoding? (https://arxiv.org/abs/2011.05315) * OpenAI's GPT-3 model (https://openai.com/api/) * OpenAI's GPT-2 model (https://openai.com/blog/tags/gpt-2/) * OpenAI's GPT-2 model (https://openai.com/blog/tags/gpt-2/) * The Part-Time Parliament, Lamport, 1998. (https://lamport.azurewebsites.net/pubs/lamport-paxos.pdf) * You Autocomplete Me: Poisoning Vulnerabilities in Neural Code Completion (https://arxiv.org/abs/2007.02220) ZK Whiteboard Sessions (https://zkhack.dev/whiteboard/) – as part of ZK Hack and powered by Polygon – a new series of educational videos that will help you get onboarded into the concepts and terms that we talk about on the ZK front. ZK Jobs Board (https://jobsboard.zeroknowledge.fm/) – has a fresh batch of open roles from ZK-focused projects. Find your next opportunity working in ZK! Today's episode is sponsored by Mina Protocol (https://minaprotocol.com/). With Mina's zero knowledge smart contracts – or zkApps – developers can create apps that offer privacy, security, and verifiability for your users. Head to minaprotocol.com/zkpodcast (http://minaprotocol.com/zkpodcast) to learn about their developer bootcamps and open grants. If you like what we do: * Find all our links here! @ZeroKnowledge | Linktree (https://linktr.ee/zeroknowledge) * Subscribe to our podcast newsletter (https://zeroknowledge.substack.com) * Follow us on Twitter @zeroknowledgefm (https://twitter.com/zeroknowledgefm) * Join us on Telegram (https://zeroknowledge.fm/telegram) * Catch us on Youtube (https://zeroknowledge.fm/) * Head to the ZK Community Forum (https://community.zeroknowledge.fm/) * Support our Gitcoin Grant (https://zeroknowledge.fm/gitcoin-grant-329-zkp-2)

Send us a textThis is the second part of my deep dive into the Lucas Research Study on Project Based Learning. You can find the first part in episode 88. The research findings absolutely dispel the myth that PBL doesn't improve test scores. Project Based Learning improves AP test scores for high school students and creates noticeable gains in middle school and elementary school students.It also increased engagement for the teachers. Having a culture with sustained professional learning and coaching for teachers and staff creates an environment where PBL is done right and increases outcomes and enjoyment for learners and teachers. The study shows that PBL was beneficial for all learners and staff. Episode Highlights: [02:10] "Need to Know" - How does our PBL school get better throughout the school year? [02:50] Having a staff member or PBL coach that specifically looks at Project Based Learning and supports the staff is essential. [04:51] Administrators also need an administrative coach. You need someone outside of your staff and district to push you.[06:04] Project Based Learning boosts student learning in AP courses.[07:48] Many of the participants were from underserved communities, and the study was created to help promote equity. The program increased test performance when using Project Based Learning.[08:31] The myth that Project Based Learning does not enhance test scores has been flipped on its head. [09:37] Your staff has to have professional learning in order to implement PBL properly.[10:24] PBL also led to gains in math, science, and vocabulary learning for middle schoolers. The teachers were also more engaged.[14:48] PBL is beneficial for all students.Resources & Links Related to this EpisodeWhat is PBL?Ask RyanMagnify Learning YouTubeProject Based Learning Stories and Structures: Wins, Fails, and Where to StartMagnify LearningRyan Steuer Twitter @ryansteuerPBL Workshops For Schools & DistrictsCommunity Partner ResourcesLucas Research Briefs on Project Based LearningPBL Leadership - Project Based Learning Research Part 1 | Episode 88PBL Guest - Danny Bauer of Better Leaders Better Schools | Episode 87

This is the second part of my deep dive into the Lucas Research Study on Project Based Learning. You can find the first part in episode 88. The research findings absolutely dispel the myth that PBL doesn't improve test scores. Project Based Learning improves AP test scores for high school students and creates noticeable gains in middle school and elementary school students.It also increased engagement for the teachers. Having a culture with sustained professional learning and coaching for teachers and staff creates an environment where PBL is done right and increases outcomes and enjoyment for learners and teachers. The study shows that PBL was beneficial for all learners and staff. Episode Highlights: [02:10] "Need to Know" - How does our PBL school get better throughout the school year? [02:50] Having a staff member or PBL coach that specifically looks at Project Based Learning and supports the staff is essential. [04:51] Administrators also need an administrative coach. You need someone outside of your staff and district to push you.[06:04] Project Based Learning boosts student learning in AP courses.[07:48] Many of the participants were from underserved communities, and the study was created to help promote equity. The program increased test performance when using Project Based Learning.[08:31] The myth that Project Based Learning does not enhance test scores has been flipped on its head. [09:37] Your staff has to have professional learning in order to implement PBL properly.[10:24] PBL also led to gains in math, science, and vocabulary learning for middle schoolers. The teachers were also more engaged.[14:48] PBL is beneficial for all students.Resources & Links Related to this EpisodeWhat is PBL?Ask RyanMagnify Learning YouTubeProject Based Learning Stories and Structures: Wins, Fails, and Where to StartMagnify LearningRyan Steuer Twitter @ryansteuerPBL Workshops For Schools & DistrictsCommunity Partner ResourcesLucas Research Briefs on Project Based LearningPBL Leadership - Project Based Learning Research Part 1 | Episode 88PBL Guest - Danny Bauer of Better Leaders Better Schools | Episode 87

I've seen lives turned around by Project Based Learning. I have loads of success stories from learners, teachers, and administrators. There is also research to back up the benefits of Project Based Learning. This leadership episode is going to be a two-part episode focusing on Research on Project Based Learning by Lucas Education Research. The brief is linked to in the resources below.This episode takes a high-level view, and then I break it down in the upcoming episode 90 as we take a deeper look. We also have a “need to know” that focuses on the question of where do I get started with PBL? I share a free resource in the show notes and other options for learning about Project Based Learning. I also share the best way to learn by visiting a school and seeing it in action. Episode Highlights: [02:54] Where do I get started with PBL? Educate yourself with books, podcasts, videos, and visiting a school that is doing PBL.[04:01] On your visit ask good questions and be sure to talk to the learners.[05:55] It's a powerful lever for improving equity. When it's used significant learning occurs. [07:05] The study shows that when underserved students engage in PBL they learn significantly.[09:22] Learners miss out on opportunities and authentic learning experiences when not involved in Project Based Learning.[11:59] Curriculum needs to be flexible enough to pull in additional resources to help your students connect it to the work. Connections are a huge benefit.[13:06] Belonging in the school community. We need to build the culture so students are in a safe environment.[15:37] PBL also creates strong learning opportunities for teachers. Sustained high-quality professional learning. Resources & Links Related to this EpisodeWhat is PBL?Ask RyanMagnify Learning YouTubeProject Based Learning Stories and Structures: Wins, Fails, and Where to StartMagnify LearningRyan Steuer Twitter @ryansteuerPBL Workshops For Schools & DistrictsCommunity Partner ResourcesLucas Research Briefs on Project Based Learning

Send us a textI've seen lives turned around by Project Based Learning. I have loads of success stories from learners, teachers, and administrators. There is also research to back up the benefits of Project Based Learning. This leadership episode is going to be a two-part episode focusing on Research on Project Based Learning by Lucas Education Research. The brief is linked to in the resources below.This episode takes a high-level view, and then I break it down in the upcoming episode 90 as we take a deeper look. We also have a “need to know” that focuses on the question of where do I get started with PBL? I share a free resource in the show notes and other options for learning about Project Based Learning. I also share the best way to learn by visiting a school and seeing it in action. Episode Highlights: [02:54] Where do I get started with PBL? Educate yourself with books, podcasts, videos, and visiting a school that is doing PBL.[04:01] On your visit ask good questions and be sure to talk to the learners.[05:55] It's a powerful lever for improving equity. When it's used significant learning occurs. [07:05] The study shows that when underserved students engage in PBL they learn significantly.[09:22] Learners miss out on opportunities and authentic learning experiences when not involved in Project Based Learning.[11:59] Curriculum needs to be flexible enough to pull in additional resources to help your students connect it to the work. Connections are a huge benefit.[13:06] Belonging in the school community. We need to build the culture so students are in a safe environment.[15:37] PBL also creates strong learning opportunities for teachers. Sustained high-quality professional learning. Resources & Links Related to this EpisodeWhat is PBL?Ask RyanMagnify Learning YouTubeProject Based Learning Stories and Structures: Wins, Fails, and Where to StartMagnify LearningRyan Steuer Twitter @ryansteuerPBL Workshops For Schools & DistrictsCommunity Partner ResourcesLucas Research Briefs on Project Based Learning

If you have a child (a young sibling, cousin, student, or even friend) in your life, chances are you know about Scratch—the wildly popular graphical program language that kids use to dream up interactive stories, games, and animations. But it's entirely possible you don't know the man behind Scratch, Mitch Resnick, the LEGO Papert Professor of Learning Research at the MIT Media Lab, or that the Scratch Foundation has a strong, long-term relationship with EPAM. After listening to the latest iteration of *The Resonance Test,* in which Resnick and Shamilka Samarasinha, EPAM's Global Head of Corporate Social Responsibility, answer questions from producer Ken Gordon, that ignorance will instantly evaporate. The episode digs into the reasons why Scratch is, and always has been, a free program (“We didn't want there to be barriers for young people to get access to Scratch,” says Resnick) and how Scratch helped children during Covid (the first year of the pandemic saw the number of Scratch projects double and the number of comments the kids wrote on each other's projects increase fivefold). You'll hear about Scratch and the kids of Ukraine. Says Resnick: “In early March, 10 days after the invasion of Ukraine, I got a message from an educator in Ukraine named Olesia Vlasii.” Vlasii had the idea to use Scratch to create what she called Waves of Kindness. The result: A Waves of Kindness gallery on the Scratch website “where kids from around the world could upload projects about how you could spread kindness,” says Resnick. Within days, Waves of Kindness featured “literally thousands of projects from kids around the world.” The conversation also touches on how Scratch engages a wide ecosystem of learners to promote diversity and inclusion in expanding education and how EPAM's partnership with Scratch fits into our other ESG activities. “In the social impact space, obviously education is one of our core areas,” says Samarasinha. Finally, Resnik and Samarasinha talk about the evolving relationship between the Scratch Foundation and EPAM—our EPAM E-Kids program has expanded from four to 19 countries—and the upcoming virtual Scratch Conference. It's a conversation that your kid will want you to hear. So listen! Host: Alison Kotin Engineer: Kyp Pilalas Producer: Ken Gordon

Vin Vashishta is a chief data officer and AI strategist at V Squared, a company he founded in 2012 that provides AI strategy, transformation, and data organizational build-out services. He teaches data professionals about strategy, communications, business acumen, and applied machine learning research methods. Vin has 130k+ followers on Linkedin talking about AI, analytics, and strategy. His website: https://www.datascience.vin/ Follow @DalianaLiu for more updates on data science and this show. If you enjoy this episode, subscribe and leave a 5-star review :) Topics: Machine learning problem solving Case study: how to find pricing strategy through data science Applied ML research methods “Human machine team” Casual inference resources/books How to earn trust from customer Future of data science/Machine learning MLOps vs QA (quality control) How to lead without authority Mistakes he made in his career What he learned from his mentor Shift in data science over the passed 10 years

Adam and Kathryn talk about Pilates, yoga and how we can integrate current evidence into the practices we love without throwing the baby out with the bath water. Adam is on his way to becoming a physical therapist and along the path has incorporated heavy weights into his long-time Pilates career. They also talk about motor learning and how we can cue our clients to be as successful as possible during exercise. ___ If this conversation excites you click here to learn more about the Mindful Strength Teacher's Immersion 2.0. Early bird ends July 1st. This 3-month continuing education course for yoga, pilates and fitness teachers begins in October. ___ Adam lives in Long Beach, California and has been teaching Pilates since 2009. He is an internationally recognized Pilates teacher trainer for Breathe Education, holds a Bachelors of Science in Exercise Science and is currently studying Physical Therapy at the University of St. Augustine for Health Sciences. Adam is the founder of various live online workshops such as “Motor Learning Strategies” & “Pain Science & Pilates To learn more about Adam click here. To follow Adam on IG click here.  

#dsi #search #google Search engines work by building an index and then looking up things in it. Usually, that index is a separate data structure. In keyword search, we build and store reverse indices. In neural search, we build nearest-neighbor indices. This paper does something different: It directly trains a Transformer to return the ID of the most relevant document. No similarity search over embeddings or anything like this is performed, and no external data structure is needed, as the entire index is essentially captured by the model's weights. The paper experiments with various ways of representing documents and training the system, which works surprisingly well! Sponsor: Diffgram https://diffgram.com?ref=yannic OUTLINE: 0:00 - Intro 0:45 - Sponsor: Diffgram 1:35 - Paper overview 3:15 - The search problem, classic and neural 8:15 - Seq2seq for directly predicting document IDs 11:05 - Differentiable search index architecture 18:05 - Indexing 25:15 - Retrieval and document representation 33:25 - Training DSI 39:15 - Experimental results 49:25 - Comments & Conclusions Paper: https://arxiv.org/abs/2202.06991 Abstract: In this paper, we demonstrate that information retrieval can be accomplished with a single Transformer, in which all information about the corpus is encoded in the parameters of the model. To this end, we introduce the Differentiable Search Index (DSI), a new paradigm that learns a text-to-text model that maps string queries directly to relevant docids; in other words, a DSI model answers queries directly using only its parameters, dramatically simplifying the whole retrieval process. We study variations in how documents and their identifiers are represented, variations in training procedures, and the interplay between models and corpus sizes. Experiments demonstrate that given appropriate design choices, DSI significantly outperforms strong baselines such as dual encoder models. Moreover, DSI demonstrates strong generalization capabilities, outperforming a BM25 baseline in a zero-shot setup. Authors: Yi Tay, Vinh Q. Tran, Mostafa Dehghani, Jianmo Ni, Dara Bahri, Harsh Mehta, Zhen Qin, Kai Hui, Zhe Zhao, Jai Gupta, Tal Schuster, William W. Cohen, Donald Metzler Links: Merch: store.ykilcher.com TabNine Code Completion (Referral): http://bit.ly/tabnine-yannick YouTube: https://www.youtube.com/c/yannickilcher Twitter: https://twitter.com/ykilcher Discord: https://discord.gg/4H8xxDF BitChute: https://www.bitchute.com/channel/yann... LinkedIn: https://www.linkedin.com/in/ykilcher BiliBili: https://space.bilibili.com/2017636191 If you want to support me, the best thing to do is to share out the content :) If you want to support me financially (completely optional and voluntary, but a lot of people have asked for this): SubscribeStar: https://www.subscribestar.com/yannick... Patreon: https://www.patreon.com/yannickilcher Bitcoin (BTC): bc1q49lsw3q325tr58ygf8sudx2dqfguclvngvy2cq Ethereum (ETH): 0x7ad3513E3B8f66799f507Aa7874b1B0eBC7F85e2 Litecoin (LTC): LQW2TRyKYetVC8WjFkhpPhtpbDM4Vw7r9m Monero (XMR): 4ACL8AGrEo5hAir8A9CeVrW8pEauWvnp1WnSDZxW7tziCDLhZAGsgzhRQABDnFy8yuM9fWJDviJPHKRjV4FWt19CJZN9D4n

Today I had the pleasure of interviewing Yannic Kilcher. Yannic is a YouTuber covering state of the art Machine Learning research topics. He has a PhD from ETH Zurich and is currently the CTO of DeepJudge, a LegalTech NLP startup. In this episode we learn about how Yannic decided on a PHD in Ai, how he is able to make advanced research so digestable, and the reason why he wears sunglasses on camera. I hope you enjoy the epsisode, I know I enjoyed our conversation.

“Why is the sky blue?” “Why do people get sick?” “Why aren't there any more dinosaurs?” Sometimes it feels like children never stop asking questions. And they shouldn't. A recent OECD International Early Learning and Child Wellbeing study shows that children who are curious have stronger language and number skills, and better self-control. So how do we keep students curious and creative even after they've outgrown kindergarten? Rowena Phair, senior analyst at the OECD, and Mitch Resnick, Professor of Learning Research at the MIT Media Lab, discuss. Host: Clara Young; Producer: Ilse Sánchez

Mathilde Caron is a PhD. candidate at the French National Institute for Research in Digital Science and Technology and at Facebook AI (Meta AI). She does the majority of her research in the field of Machine learning called self-supervised learning. She has a few first authorships on important academic papers in the space. Her work: https://scholar.google.com/citations?user=eiB0s-kAAAAJ&hl=fr You can donate to this podcast at this bitcoin address: 33wejXuGGDtQj9GPwCgjwPxPq4dc4muZjg --- This episode is sponsored by · Anchor: The easiest way to make a podcast. https://anchor.fm/app Support this podcast: https://anchor.fm/idris-sunmola/support

#grafting #adam #sgd The last years in deep learning research have given rise to a plethora of different optimization algorithms, such as SGD, AdaGrad, Adam, LARS, LAMB, etc. which all claim to have their special peculiarities and advantages. In general, all algorithms modify two major things: The (implicit) learning rate schedule, and a correction to the gradient direction. This paper introduces grafting, which allows to transfer the induced learning rate schedule of one optimizer to another one. In that, the paper shows that much of the benefits of adaptive methods (e.g. Adam) are actually due to this schedule, and not necessarily to the gradient direction correction. Grafting allows for more fundamental research into differences and commonalities between optimizers, and a derived version of it makes it possible to computes static learning rate corrections for SGD, which potentially allows for large savings of GPU memory. OUTLINE 0:00 - Rant about Reviewer #2 6:25 - Intro & Overview 12:25 - Adaptive Optimization Methods 20:15 - Grafting Algorithm 26:45 - Experimental Results 31:35 - Static Transfer of Learning Rate Ratios 35:25 - Conclusion & Discussion Paper (OpenReview): https://openreview.net/forum?id=FpKgG... Old Paper (Arxiv): https://arxiv.org/abs/2002.11803 Our Discord: https://discord.gg/4H8xxDF Abstract: In the empirical science of training large neural networks, the learning rate schedule is a notoriously challenging-to-tune hyperparameter, which can depend on all other properties (architecture, optimizer, batch size, dataset, regularization, ...) of the problem. In this work, we probe the entanglements between the optimizer and the learning rate schedule. We propose the technique of optimizer grafting, which allows for the transfer of the overall implicit step size schedule from a tuned optimizer to a new optimizer, preserving empirical performance. This provides a robust plug-and-play baseline for optimizer comparisons, leading to reductions to the computational cost of optimizer hyperparameter search. Using grafting, we discover a non-adaptive learning rate correction to SGD which allows it to train a BERT model to state-of-the-art performance. Besides providing a resource-saving tool for practitioners, the invariances discovered via grafting shed light on the successes and failure modes of optimizers in deep learning. Authors: Anonymous (Under Review) Links: TabNine Code Completion (Referral): http://bit.ly/tabnine-yannick YouTube: https://www.youtube.com/c/yannickilcher Twitter: https://twitter.com/ykilcher Discord: https://discord.gg/4H8xxDF BitChute: https://www.bitchute.com/channel/yann... LinkedIn: https://www.linkedin.com/in/ykilcher BiliBili: https://space.bilibili.com/2017636191 If you want to support me, the best thing to do is to share out the content :) If you want to support me financially (completely optional and voluntary, but a lot of people have asked for this): SubscribeStar: https://www.subscribestar.com/yannick... Patreon: https://www.patreon.com/yannickilcher Bitcoin (BTC): bc1q49lsw3q325tr58ygf8sudx2dqfguclvngvy2cq Ethereum (ETH): 0x7ad3513E3B8f66799f507Aa7874b1B0eBC7F85e2 Litecoin (LTC): LQW2TRyKYetVC8WjFkhpPhtpbDM4Vw7r9m Monero (XMR): 4ACL8AGrEo5hAir8A9CeVrW8pEauWvnp1WnSDZxW7tziCDLhZAGsgzhRQABDnFy8yuM9fWJDviJPHKRjV4FWt19CJZN9D4n

#deeplearning #backpropagation #simulation More and more systems are made differentiable, which means that accurate gradients of these systems' dynamics can be computed exactly. While this development has led to a lot of advances, there are also distinct situations where backpropagation can be a very bad idea. This paper characterizes a few such systems in the domain of iterated dynamical systems, often including some source of stochasticity, resulting in chaotic behavior. In these systems, it is often better to use black-box estimators for gradients than computing them exactly. OUTLINE: 0:00 - Foreword 1:15 - Intro & Overview 3:40 - Backpropagation through iterated systems 12:10 - Connection to the spectrum of the Jacobian 15:35 - The Reparameterization Trick 21:30 - Problems of reparameterization 26:35 - Example 1: Policy Learning in Simulation 33:05 - Example 2: Meta-Learning Optimizers 36:15 - Example 3: Disk packing 37:45 - Analysis of Jacobians 40:20 - What can be done? 45:40 - Just use Black-Box methods Paper: https://arxiv.org/abs/2111.05803 Abstract: Differentiable programming techniques are widely used in the community and are responsible for the machine learning renaissance of the past several decades. While these methods are powerful, they have limits. In this short report, we discuss a common chaos based failure mode which appears in a variety of differentiable circumstances, ranging from recurrent neural networks and numerical physics simulation to training learned optimizers. We trace this failure to the spectrum of the Jacobian of the system under study, and provide criteria for when a practitioner might expect this failure to spoil their differentiation based optimization algorithms. Authors: Luke Metz, C. Daniel Freeman, Samuel S. Schoenholz, Tal Kachman Links: TabNine Code Completion (Referral): http://bit.ly/tabnine-yannick YouTube: https://www.youtube.com/c/yannickilcher Twitter: https://twitter.com/ykilcher Discord: https://discord.gg/4H8xxDF BitChute: https://www.bitchute.com/channel/yann... LinkedIn: https://www.linkedin.com/in/ykilcher BiliBili: https://space.bilibili.com/2017636191 If you want to support me, the best thing to do is to share out the content :) If you want to support me financially (completely optional and voluntary, but a lot of people have asked for this): SubscribeStar: https://www.subscribestar.com/yannick... Patreon: https://www.patreon.com/yannickilcher Bitcoin (BTC): bc1q49lsw3q325tr58ygf8sudx2dqfguclvngvy2cq Ethereum (ETH): 0x7ad3513E3B8f66799f507Aa7874b1B0eBC7F85e2 Litecoin (LTC): LQW2TRyKYetVC8WjFkhpPhtpbDM4Vw7r9m Monero (XMR): 4ACL8AGrEo5hAir8A9CeVrW8pEauWvnp1WnSDZxW7tziCDLhZAGsgzhRQABDnFy8yuM9fWJDviJPHKRjV4FWt19CJZN9D4n

#machinelearning #ardm #generativemodels Diffusion models have made large advances in recent months as a new type of generative models. This paper introduces Autoregressive Diffusion Models (ARDMs), which are a mix between autoregressive generative models and diffusion models. ARDMs are trained to be agnostic to the order of autoregressive decoding and give the user a dynamic tradeoff between speed and performance at decoding time. This paper applies ARDMs to both text and image data, and as an extension, the models can also be used to perform lossless compression. OUTLINE: 0:00 - Intro & Overview 3:15 - Decoding Order in Autoregressive Models 6:15 - Autoregressive Diffusion Models 8:35 - Dependent and Independent Sampling 14:25 - Application to Character-Level Language Models 18:15 - How Sampling & Training Works 26:05 - Extension 1: Parallel Sampling 29:20 - Extension 2: Depth Upscaling 33:10 - Conclusion & Comments Paper: https://arxiv.org/abs/2110.02037 Abstract: We introduce Autoregressive Diffusion Models (ARDMs), a model class encompassing and generalizing order-agnostic autoregressive models (Uria et al., 2014) and absorbing discrete diffusion (Austin et al., 2021), which we show are special cases of ARDMs under mild assumptions. ARDMs are simple to implement and easy to train. Unlike standard ARMs, they do not require causal masking of model representations, and can be trained using an efficient objective similar to modern probabilistic diffusion models that scales favourably to highly-dimensional data. At test time, ARDMs support parallel generation which can be adapted to fit any given generation budget. We find that ARDMs require significantly fewer steps than discrete diffusion models to attain the same performance. Finally, we apply ARDMs to lossless compression, and show that they are uniquely suited to this task. Contrary to existing approaches based on bits-back coding, ARDMs obtain compelling results not only on complete datasets, but also on compressing single data points. Moreover, this can be done using a modest number of network calls for (de)compression due to the model's adaptable parallel generation. Authors: Emiel Hoogeboom, Alexey A. Gritsenko, Jasmijn Bastings, Ben Poole, Rianne van den Berg, Tim Salimans Links: TabNine Code Completion (Referral): http://bit.ly/tabnine-yannick YouTube: https://www.youtube.com/c/yannickilcher Twitter: https://twitter.com/ykilcher Discord: https://discord.gg/4H8xxDF BitChute: https://www.bitchute.com/channel/yann... Minds: https://www.minds.com/ykilcher Parler: https://parler.com/profile/YannicKilcher LinkedIn: https://www.linkedin.com/in/ykilcher BiliBili: https://space.bilibili.com/1824646584 If you want to support me, the best thing to do is to share out the content :) If you want to support me financially (completely optional and voluntary, but a lot of people have asked for this): SubscribeStar: https://www.subscribestar.com/yannick... Patreon: https://www.patreon.com/yannickilcher Bitcoin (BTC): bc1q49lsw3q325tr58ygf8sudx2dqfguclvngvy2cq Ethereum (ETH): 0x7ad3513E3B8f66799f507Aa7874b1B0eBC7F85e2 Litecoin (LTC): LQW2TRyKYetVC8WjFkhpPhtpbDM4Vw7r9m

#efficientzero #muzero #atari Reinforcement Learning methods are notoriously data-hungry. Notably, MuZero learns a latent world model just from scalar feedback of reward- and policy-predictions, and therefore relies on scale to perform well. However, most RL algorithms fail when presented with very little data. EfficientZero makes several improvements over MuZero that allows it to learn from astonishingly small amounts of data and outperform other methods by a large margin in the low-sample setting. This could be a staple algorithm for future RL research. OUTLINE: 0:00 - Intro & Outline 2:30 - MuZero Recap 10:50 - EfficientZero improvements 14:15 - Self-Supervised consistency loss 17:50 - End-to-end prediction of the value prefix 20:40 - Model-based off-policy correction 25:45 - Experimental Results & Conclusion Paper: https://arxiv.org/abs/2111.00210 Code: https://github.com/YeWR/EfficientZero Note: code not there yet as of release of this video Abstract: Reinforcement learning has achieved great success in many applications. However, sample efficiency remains a key challenge, with prominent methods requiring millions (or even billions) of environment steps to train. Recently, there has been significant progress in sample efficient image-based RL algorithms; however, consistent human-level performance on the Atari game benchmark remains an elusive goal. We propose a sample efficient model-based visual RL algorithm built on MuZero, which we name EfficientZero. Our method achieves 190.4% mean human performance and 116.0% median performance on the Atari 100k benchmark with only two hours of real-time game experience and outperforms the state SAC in some tasks on the DMControl 100k benchmark. This is the first time an algorithm achieves super-human performance on Atari games with such little data. EfficientZero's performance is also close to DQN's performance at 200 million frames while we consume 500 times less data. EfficientZero's low sample complexity and high performance can bring RL closer to real-world applicability. We implement our algorithm in an easy-to-understand manner and it is available at this https URL. We hope it will accelerate the research of MCTS-based RL algorithms in the wider community. Authors: Weirui Ye, Shaohuai Liu, Thanard Kurutach, Pieter Abbeel, Yang Gao Links: TabNine Code Completion (Referral): http://bit.ly/tabnine-yannick YouTube: https://www.youtube.com/c/yannickilcher Twitter: https://twitter.com/ykilcher Discord: https://discord.gg/4H8xxDF BitChute: https://www.bitchute.com/channel/yann... Minds: https://www.minds.com/ykilcher Parler: https://parler.com/profile/YannicKilcher LinkedIn: https://www.linkedin.com/in/ykilcher BiliBili: https://space.bilibili.com/1824646584 If you want to support me, the best thing to do is to share out the content :) If you want to support me financially (completely optional and voluntary, but a lot of people have asked for this): SubscribeStar: https://www.subscribestar.com/yannick... Patreon: https://www.patreon.com/yannickilcher Bitcoin (BTC): bc1q49lsw3q325tr58ygf8sudx2dqfguclvngvy2cq Ethereum (ETH): 0x7ad3513E3B8f66799f507Aa7874b1B0eBC7F85e2 Litecoin (LTC): LQW2TRyKYetVC8WjFkhpPhtpbDM4Vw7r9m Monero (XMR): 4ACL8AGrEo5hAir8A9CeVrW8pEauWvnp1WnSDZxW7tziCDLhZAGsgzhRQABDnFy8yuM9fWJDviJPHKRjV4FWt19CJZN9D4n

BIO: Andre Hsu is a thinker and business strategist based in Singapore. He is the author of three books about qualities, mindsets, and frameworks relevant to business, which he observed in several business tycoons who left a deep and lasting impact on his life. STORY: Andre partnered with a software company with an excellent business idea, but the partners were poor in managing the company and selling the product, so it failed. Andre lost his entire investment. LEARNING: Research the people who own a business as much as you research the business.   “No matter how great the idea is, you'll not succeed if you cannot trust your partner.”Andre Hsu  Guest profilehttps://www.linkedin.com/in/andrehsujs/ (Andre Hsu) is a thinker and business strategist based in Singapore. He started his entrepreneurial journey at 17, working on a real estate project while juggling three academic degrees completed concurrently in Australia. He is the author of three books about qualities, mindsets, and frameworks that are relevant for business which he observed in several business tycoons who left a deep and lasting impact on his life. Andre likes to use multiple techniques to read, predict people, assess situations, and formulate strategies that are suitable for properties and negotiations, deal structuring related to the assets. He likes to educate, share knowledge, insights, and reasoning of strategies to business associates who then proceed to implement them. He looks forward to doing this with people who share similar values and visions. Worst investment everAfter finishing university, Andre went into the family business, and after a while, he decided to venture into his own business pursuits. He got involved in a small software company that was dealing with Point-of-Service systems. At the time, this was a very lucrative business because not many companies were using POS systems. Andre was, therefore, happy to partner with the company and invest in this venture. The mistake Andre made was investing in people who didn't take the business part of the venture seriously. They only created a good product, but they never invested in sales or management, so the product never really took off. Lessons learnedYou need to research the people who own a business as much as you research the business. Andrew's takeawaysWhen investing in a startup, you need to look for trust, a good idea, the ability to execute the idea, and capital. Actionable adviceIf you cannot trust the person you want to partner with, forget the idea. No matter how great the idea is, you'll not succeed if you cannot trust your partner. No. 1 goal for the next 12 monthsAndre Hsu's number one goal for the next 12 months is to step back from the business and have his partners run it to have more time to do strategic thinking and come up with new ideas.   [spp-transcript]   Connect with Andre Hsuhttps://www.linkedin.com/in/andrehsujs/ (LinkedIn) https://amzn.to/3nNgf64 (Book) Andrew's bookshttps://amzn.to/3qrfHjX (How to Start Building Your Wealth Investing in the Stock Market) https://amzn.to/2PDApAo (My Worst Investment Ever) https://amzn.to/3v6ip1Y (9 Valuation Mistakes and How to Avoid Them) https://amzn.to/3emBO8M (Transform Your Business with Dr.Deming's 14 Points) Andrew's online programshttps://valuationmasterclass.com/ (Valuation Master Class) https://academy.astotz.com/courses/how-to-start-building-your-wealth-investing-in-the-stock-market (How to Start Building Your Wealth Investing in the Stock Market) https://academy.astotz.com/courses/finance-made-ridiculously-simple (Finance Made Ridiculously Simple) https://academy.astotz.com/courses/gp (Become a Great Presenter and Increase Your Influence) https://academy.astotz.com/courses/transformyourbusiness (Transform Your Business with Dr. Deming's 14 Points) Connect with Andrew Stotz:https://www.astotz.com/ (astotz.com) https://www.linkedin.com/in/andrewstotz/ (LinkedIn)...

#tvae #topographic #equivariant Variational Autoencoders model the latent space as a set of independent Gaussian random variables, which the decoder maps to a data distribution. However, this independence is not always desired, for example when dealing with video sequences, we know that successive frames are heavily correlated. Thus, any latent space dealing with such data should reflect this in its structure. Topographic VAEs are a framework for defining correlation structures among the latent variables and induce equivariance within the resulting model. This paper shows how such correlation structures can be built by correctly arranging higher-level variables, which are themselves independent Gaussians. OUTLINE: 0:00 - Intro 1:40 - Architecture Overview 6:30 - Comparison to regular VAEs 8:35 - Generative Mechanism Formulation 11:45 - Non-Gaussian Latent Space 17:30 - Topographic Product of Student-t 21:15 - Introducing Temporal Coherence 24:50 - Topographic VAE 27:50 - Experimental Results 31:15 - Conclusion & Comments Paper: https://arxiv.org/abs/2109.01394 Code: https://github.com/akandykeller/topog... Abstract: In this work we seek to bridge the concepts of topographic organization and equivariance in neural networks. To accomplish this, we introduce the Topographic VAE: a novel method for efficiently training deep generative models with topographically organized latent variables. We show that such a model indeed learns to organize its activations according to salient characteristics such as digit class, width, and style on MNIST. Furthermore, through topographic organization over time (i.e. temporal coherence), we demonstrate how predefined latent space transformation operators can be encouraged for observed transformed input sequences -- a primitive form of unsupervised learned equivariance. We demonstrate that this model successfully learns sets of approximately equivariant features (i.e. "capsules") directly from sequences and achieves higher likelihood on correspondingly transforming test sequences. Equivariance is verified quantitatively by measuring the approximate commutativity of the inference network and the sequence transformations. Finally, we demonstrate approximate equivariance to complex transformations, expanding upon the capabilities of existing group equivariant neural networks. Authors: T. Anderson Keller, Max Welling Links: TabNine Code Completion (Referral): http://bit.ly/tabnine-yannick YouTube: https://www.youtube.com/c/yannickilcher Twitter: https://twitter.com/ykilcher Discord: https://discord.gg/4H8xxDF BitChute: https://www.bitchute.com/channel/yann... Minds: https://www.minds.com/ykilcher Parler: https://parler.com/profile/YannicKilcher LinkedIn: https://www.linkedin.com/in/ykilcher BiliBili: https://space.bilibili.com/1824646584 If you want to support me, the best thing to do is to share out the content :) If you want to support me financially (completely optional and voluntary, but a lot of people have asked for this): SubscribeStar: https://www.subscribestar.com/yannick... Patreon: https://www.patreon.com/yannickilcher Bitcoin (BTC): bc1q49lsw3q325tr58ygf8sudx2dqfguclvngvy2cq Ethereum (ETH): 0x7ad3513E3B8f66799f507Aa7874b1B0eBC7F85e2 Litecoin (LTC): LQW2TRyKYetVC8WjFkhpPhtpbDM4Vw7r9m Monero (XMR): 4ACL8AGrEo5hAir8A9CeVrW8pEauWvnp1WnSDZxW7tziCDLhZAGsgzhRQABDnFy8yuM9fWJDviJPHKRjV4FWt19CJZN9D4n

#attention #transformer #fastformer Transformers have become the dominant model class in the last few years for large data, but their quadratic complexity in terms of sequence length has plagued them until now. Fastformer claims to be the fastest and most performant linear attention variant, able to consume long contexts at once. This is achieved by a combination of additive attention and elementwise products. While initial results look promising, I have my reservations... OUTLINE: 0:00 - Intro & Outline 2:15 - Fastformer description 5:20 - Baseline: Classic Attention 10:00 - Fastformer architecture 12:50 - Additive Attention 18:05 - Query-Key element-wise multiplication 21:35 - Redundant modules in Fastformer 25:00 - Problems with the architecture 27:30 - Is this even attention? 32:20 - Experimental Results 34:50 - Conclusion & Comments Paper: https://arxiv.org/abs/2108.09084 Abstract: Transformer is a powerful model for text understanding. However, it is inefficient due to its quadratic complexity to input sequence length. Although there are many methods on Transformer acceleration, they are still either inefficient on long sequences or not effective enough. In this paper, we propose Fastformer, which is an efficient Transformer model based on additive attention. In Fastformer, instead of modeling the pair-wise interactions between tokens, we first use additive attention mechanism to model global contexts, and then further transform each token representation based on its interaction with global context representations. In this way, Fastformer can achieve effective context modeling with linear complexity. Extensive experiments on five datasets show that Fastformer is much more efficient than many existing Transformer models and can meanwhile achieve comparable or even better long text modeling performance. Authors: Chuhan Wu, Fangzhao Wu, Tao Qi, Yongfeng Huang Links: TabNine Code Completion (Referral): http://bit.ly/tabnine-yannick YouTube: https://www.youtube.com/c/yannickilcher Twitter: https://twitter.com/ykilcher Discord: https://discord.gg/4H8xxDF BitChute: https://www.bitchute.com/channel/yann... Minds: https://www.minds.com/ykilcher Parler: https://parler.com/profile/YannicKilcher LinkedIn: https://www.linkedin.com/in/yannic-ki... BiliBili: https://space.bilibili.com/1824646584 If you want to support me, the best thing to do is to share out the content :) If you want to support me financially (completely optional and voluntary, but a lot of people have asked for this): SubscribeStar: https://www.subscribestar.com/yannick... Patreon: https://www.patreon.com/yannickilcher Bitcoin (BTC): bc1q49lsw3q325tr58ygf8sudx2dqfguclvngvy2cq Ethereum (ETH): 0x7ad3513E3B8f66799f507Aa7874b1B0eBC7F85e2 Litecoin (LTC): LQW2TRyKYetVC8WjFkhpPhtpbDM4Vw7r9m Monero (XMR): 4ACL8AGrEo5hAir8A9CeVrW8pEauWvnp1WnSDZxW7tziCDLhZAGsgzhRQABDnFy8yuM9fWJDviJPHKRjV4FWt19CJZN9D4n

#pondernet #deepmind #machinelearning Humans don't spend the same amount of mental effort on all problems equally. Instead, we respond quickly to easy tasks, and we take our time to deliberate hard tasks. DeepMind's PonderNet attempts to achieve the same by dynamically deciding how many computation steps to allocate to any single input sample. This is done via a recurrent architecture and a trainable function that computes a halting probability. The resulting model performs well in dynamic computation tasks and is surprisingly robust to different hyperparameter settings. OUTLINE: 0:00 - Intro & Overview 2:30 - Problem Statement 8:00 - Probabilistic formulation of dynamic halting 14:40 - Training via unrolling 22:30 - Loss function and regularization of the halting distribution 27:35 - Experimental Results 37:10 - Sensitivity to hyperparameter choice 41:15 - Discussion, Conclusion, Broader Impact Paper: https://arxiv.org/abs/2107.05407 Abstract: In standard neural networks the amount of computation used grows with the size of the inputs, but not with the complexity of the problem being learnt. To overcome this limitation we introduce PonderNet, a new algorithm that learns to adapt the amount of computation based on the complexity of the problem at hand. PonderNet learns end-to-end the number of computational steps to achieve an effective compromise between training prediction accuracy, computational cost and generalization. On a complex synthetic problem, PonderNet dramatically improves performance over previous adaptive computation methods and additionally succeeds at extrapolation tests where traditional neural networks fail. Also, our method matched the current state of the art results on a real world question and answering dataset, but using less compute. Finally, PonderNet reached state of the art results on a complex task designed to test the reasoning capabilities of neural networks.1 Authors: Andrea Banino, Jan Balaguer, Charles Blundell Links: TabNine Code Completion (Referral): http://bit.ly/tabnine-yannick YouTube: https://www.youtube.com/c/yannickilcher Twitter: https://twitter.com/ykilcher Discord: https://discord.gg/4H8xxDF BitChute: https://www.bitchute.com/channel/yann... Minds: https://www.minds.com/ykilcher Parler: https://parler.com/profile/YannicKilcher LinkedIn: https://www.linkedin.com/in/yannic-ki... BiliBili: https://space.bilibili.com/1824646584 If you want to support me, the best thing to do is to share out the content :) If you want to support me financially (completely optional and voluntary, but a lot of people have asked for this): SubscribeStar: https://www.subscribestar.com/yannick... Patreon: https://www.patreon.com/yannickilcher Bitcoin (BTC): bc1q49lsw3q325tr58ygf8sudx2dqfguclvngvy2cq Ethereum (ETH): 0x7ad3513E3B8f66799f507Aa7874b1B0eBC7F85e2 Litecoin (LTC): LQW2TRyKYetVC8WjFkhpPhtpbDM4Vw7r9m Monero (XMR): 4ACL8AGrEo5hAir8A9CeVrW8pEauWvnp1WnSDZxW7tziCDLhZAGsgzhRQABDnFy8yuM9fWJDviJPHKRjV4FWt19CJZN9D4n

During this past year, one of the biggest challenges educators reported was keeping students engaged and motivated while learning online during the pandemic. In this episode of BRIGHT, we talk to Dr. Chris Harrington, the director of the Michigan Virtual Learning Research Institute, who shares his reflections on pandemic teaching, common misconceptions about online learning, and the findings of his team's landmark study on student engagement.

In this episode of the Data Exchange, our special correspondent and managing editor Jenn Webb organized a mini-panel composed of myself and Simon Rodriguez,  Data Research Assistant at the Center for Security and Emerging Technology (CSET) at Georgetown University.  Through a series of reports and data briefs, CSET provides policymakers with data rich material to inform and guide public policy.Subscribe: Apple • Android • Spotify • Stitcher • Google • RSS.Detailed show notes can be found on The Data Exchange web site.Subscribe to The Gradient Flow Newsletter.

Today, we are joined by Dr. Louis Gomez, senior fellow at the Carnegie Foundation for the Advancement of Teaching and professor of education and information studies at UCLA. Dr. Gomez addresses the topics of Networked Improvement Communities (NICs) and the mindset shift that is necessary in order to promote equity in education. Join our conversation about how compliance can prevent initiatives from being implemented, the necessity of having a common aim and narrative when discussing improvement science as part of NICs, and that equity without the will to change or the respect for the community will not bring about social justice. Visit our Bronx ART website and connect with us on Twitter @BX_ARTeam! Today's hosts are Kris DeFilippis, Adelia Gibson, and Kaitlyn Reilley Guest Information: Dr. Gomez earned a bachelor's degree in psychology from Stonybrook University and a doctorate in Psychology from University of California, Berkeley. He spent 14 years working in cognitive science and person–computer systems interactions at Bell Laboratories, Bell Communications Research Inc. and Bellcore. Dr. Gomez has held a number of faculty positions including positions at Northwestern University and the University of Pittsburgh, where he was also director of the Center for Urban Education and a senior scientist at the Learning Research and Development Center. Dr. Gomez is currently a professor of education and information studies at the University of California, Los Angeles. Since 2008, he has served as a senior fellow at the Carnegie Foundation for the Advancement of Teaching, where he leads the Network Development work. He is the co-author of Learning to Improve: How America's Schools Can Get Better at Getting Better. Dr. Gomez is dedicated to educational improvement and his numerous publications and studies have contributed greatly to bringing improvement science to the field of education. Connect with Dr. Gomez through email at lmgomez@ucla.edu Resources for Listeners: Information about iLEAD Learning to Improve: How America's Schools Can Get Better and Getting Better Why a NIC? Getting Ideas into Action: Building Networked Improvement Communities in Education Improvement Research Carried Out Through Networked Communities: Accelerating Learning about Practices that Support More Productive Student Mindsets How a Networked Improvement Community Improved Success Rates for Struggling College Math Students

GAMES IN EDUCATION PEARSON RESEARCH 's Dr Kristen DiCerbo who leads Pearson's who leads their Center for Learning Research and Technology is our guest