Localizing and Editing Knowledge in LLMs with Peter Hase - #679
Today we're joined by Peter Hase, a fifth-year PhD student at the University of North Carolina NLP lab. We discuss "scalable oversight", and the importance of developing a deeper understanding of how large neural networks make decisions. We learn how matrices are probed by interpretability researchers, and explore the two schools of thought regarding how LLMs store knowledge. Finally, we discuss the importance of deleting sensitive information from model weights, and how "easy-to-hard generalization" could increase the risk of releasing open-source foundation models.
The complete show notes for this episode can be found at twimlai.com/go/679.
Unifying Vision and Language Models with Mohit Bansal - #636
Today we're joined by Mohit Bansal, Parker Professor, and Director of the MURGe-Lab at UNC, Chapel Hill. In our conversation with Mohit, we explore the concept of unification in AI models, highlighting the advantages of shared knowledge and efficiency. He addresses the challenges of evaluation in generative AI, including biases and spurious correlations. Mohit introduces groundbreaking models such as UDOP and VL-T5, which achieved state-of-the-art results in various vision and language tasks while using fewer parameters. Finally, we discuss the importance of data efficiency, evaluating bias in models, and the future of multimodal models and explainability.
The complete show notes for this episode can be found at twimlai.com/go/636.
Artem Cherkasov and Olexandr Isayev on Democratizing Drug Discovery with Deep Learning - Ep. 172
It may seem intuitive that AI and deep learning can speed up workflows — including novel drug discovery, a typically years-long and several-billion-dollar endeavor.
But professors Artem Cherkasov and Olexandr Isayev were surprised to find that no recent academic papers provided a comprehensive, global research review of how deep learning and GPU-accelerated computing impact drug discovery.
In March, they published a paper in Nature to fill this gap, presenting an up-to-date review of the state of the art for GPU-accelerated drug discovery techniques.
Cherkasov, a professor in the department of urologic sciences at the University of British Columbia, and Isayev, an assistant professor of chemistry at Carnegie Mellon University, join NVIDIA AI Podcast host Noah Kravitz this week to discuss how GPUs can help democratize drug discovery.
In addition, the guests cover their inspiration and process for writing the paper, talk about NVIDIA technologies that are transforming the role of AI in drug discovery, and give tips for adopting new approaches to research.
Here’s How Messy a Russian Bond Default Could Be
There’s a big question over whether Russia will be able (or willing) to make payments on billions of dollars it’s borrowed from investors given its current situation. Not only does the country have a history of previous major defaults, but some of its outstanding bonds are also structured kind of strangely. On this episode of the Odd Lots podcast, Tracy Alloway and Joe Weisenthal speak with University of Virginia law professor Mitu Gulati and University of North Carolina's Mark Weidemaier. They describe how odd some Russian bonds are and what might happen after default.
See omnystudio.com/listener for privacy information.
#233 — The Groves of Misinformation
Sam Harris speaks with Zeynep Tufekci about the problem of misinformation and group-think. They discuss the Covid-19 pandemic, the early failures of journalists and public health professionals to make sense of it, the sociology of mask wearing, the problem of correcting institutional errors, Covid as a dress rehearsal for something far worse, asymmetric information warfare, failures of messaging about vaccines, the paradox of scientific authority, the power of incentives, how to reform social media, and other topics.
If the Making Sense podcast logo in your player is BLACK, you can SUBSCRIBE to gain access to all full-length episodes at samharris.org/subscribe.
From Gezi Park to Black Lives Matter, with Zeynep Tufekci
Social media-fueled protests are a force to be reckoned with in politics today. Movements like Occupy Wall Street, the Arab Spring, and Black Lives Matter have drawn millions into the streets in protest of central authorities. But can these movements be effective in the long term? Alex sits down with the field's leading writer and researcher, Zeynep Tufekci, to talk it through.
Learn more about your ad choices. Visit megaphone.fm/adchoices
Deep Learning for Population Genetic Inference with Dan Schrider - TWiML Talk #249
Today we’re joined by Dan Schrider, assistant professor in the department of genetics at UNC Chapel Hill.
My discussion with Dan starts with an overview of population genomics, looking into his application of ML in the field. We then dig into Dan’s paper “The Unreasonable Effectiveness of Convolutional Neural Networks in Population Genetic Inference,” which examines the idea that CNNs are capable of outperforming expert-derived statistical methods for some key problems in the field.
A Conversation About Go, Sci-Fi, Deep Learning and Computational Chemistry - Ep. 54
Deep learning has helped machines understand how to move pieces around a board to master, and win, Go, the most complicated game mankind has ever invented. Now it's helping a new generation of chemists better understand how to move molecules around to model new kinds of materials. Our guest, Olexandr Isayev, an assistant professor at the UNC Eshelman School of Pharmacy, at the University of North Carolina at Chapel Hill, joined our show to explain how deep learning, Go, sci-fi, and computational chemistry intersect.
We're building a dystopia just to make people click on ads | Zeynep Tufekci
We're building an artificial intelligence-powered dystopia, one click at a time, says technosociologist Zeynep Tufekci. In an eye-opening talk, she details how the same algorithms companies like Facebook, Google and Amazon use to get you to click on ads are also used to organize your access to political and social information. And the machines aren't even the real threat. What we need to understand is how the powerful might use AI to control us -- and what we can do in response.
Hosted on Acast. See acast.com/privacy for more information.
#78 — Persuasion and Control
Sam Harris speaks with Zeynep Tufekci about "surveillance capitalism," the Trump campaign's use of Facebook, AI-enabled marketing, the health of the press, Wikileaks, ransomware attacks, and other topics.
If the Making Sense podcast logo in your player is BLACK, you can SUBSCRIBE to gain access to all full-length episodes at samharris.org/subscribe.
Episode 6: Meet The Man Who Made Millions Trading Mules
This week we're thinking about what it means to be a trader in today's electronified markets and contrast it with trading in the era of horse and buggies. That's right, we're going back in time to talk mule trading and the story of the legendary Ray Lum, who spent years buying and selling livestock all over the U.S. in the early 1900s. William R. Ferris, history professor at the University of North Carolina and author of Mule Trader: Ray Lum's Tales of Horses, Mules, and Men, tells us about Lum's adventures in the South, including the purchase and transportation by train of 80,000 mules from South Dakota to New Orleans. He explains why the "trader is the poet of capitalism," how the term "day trader" can be traced to stable storage trades and why some things—like boozy dinners between brokers and their clients—never seem to change. It's arbitrage of the animal sort, with storage trades thrown in to boot.Along the way, we ask whether traders provide a social service and explore the trade-off between modern efficient markets and the bygone era of 100 percent mark-ups on (mule) trades.
See omnystudio.com/listener for privacy information.
