The Simons Foundation’s mission is to advance the frontiers of research in mathematics and the basic sciences. Co-founded in New York City by Jim and Marilyn…
3:25427 views · 18h ago · 3:25
Fireflies are an indelible part of Hillsboro, Texas. Generations of residents have spent summers catching them in jars or watching them flicker in the evening from their front porches. In this video, see how Baron and Cliett enabled their friends and neighbors to view these insects — and their hometown — in a new light.
53:27526 views · 1d ago · 53:27
Guest: Daniel Mucida
In this Presidential Lecture, Daniel Mucida will discuss surprising findings from his lab and others showing that gut-based neural circuits detect signals from infectious microbes. This process helps coordinate protective immune responses and intestinal recovery. Beyond this role in enteric neurons, his lab has begun exploring how the brain senses gut pathogens and how this sensing shapes not only sickness behavior but also immune system function.
45:29625 views · 1d ago · 45:29
Guest: Benjamin Harrop-Griffiths
Solitons and Integrable PDEs: From 19th Century Scotland to the Forefront of Contemporary Mathematics In 1834, a Scottish engineer observed a canal boat suddenly stop, creating a large, solitary wave. The seemingly simple problem of explaining this observation has led to nearly 200 years of mathematical research.
51:00350 views · 1d ago · 51:00
Guest: Emily Mackevicius
How does a collection of animals form a resilient social world? Emily Mackevicius and her team are developing a natural neuroscience of free-ranging urban rats, whose group behavior emerges from local interactions among animals navigating complex, changing environments.
49:14337 views · 1d ago · 49:14
Planets form in disks of gas and dust around young stars. Within these disks, the interactions between grain dynamics and the evolving compositions of icy grain mantles play an outsized role in regulating the bulk compositions and habitability of planets assembling in different disk locations. Karin I.
39:46276 views · 1d ago · 39:46
Guest: Carlotta Ronda
Decoding and Rewriting Microbial Worlds: From Complex Communities to Living Medicines Microbial communities are complex, dynamic systems that profoundly shape human health. Yet we still lack the ability to fully understand how their components interact and to predictably alter their collective functions.
42:52267 views · 1d ago · 42:52
Guest: Kaia Tombak
The structures of our societies have enormous impact, influencing our well-being and shaping our environments. Animal societies are structured by individual behaviors and strategies, which all depend on each other. These are sometimes based on simple rules of thumb, sometimes complex decision trees blurred by emotional noise, and are always under negotiation and adjustment as conditions change.
57:43349 views · 1d ago · 57:43
Guest: Andrew Saxe
Modern foundation models are trained through multistage pipelines, yet basic questions remain poorly understood: how should limited compute be allocated across stages, and can reinforcement learning create new capabilities during post-training or only reweight existing primitives? In this talk, Andrew Saxe will address these questions with analytical and empirical methods.
54:49358 views · 1d ago · 54:49
Guest: Surya Ganguli
Artificial intelligence has made tremendous engineering advances that stand poised to transform our society, yet alarmingly, we understand very little scientifically about how AI actually works. The time is now ripe to study AI as a complex system, by both exploiting and extending the power of physics and mathematics to open the black box of AI systems and improve them in a principled manner.
1:08:24339 views · 1d ago · 1:08:24
Guest: Yasaman Bahri
Learned representations lie at the core of AI systems, connecting the statistics of data to nontrivial computation. What ingredients shape these representations, and what can we learn by studying their geometry? How do representations enable generalization and the emergence of new computational capabilities?
46:16419 views · 1d ago · 46:16
Guest: Michael Douglas
Over the past year, AI systems have crossed into research-level mathematics — solving competition and open problems whose solutions can be checked exactly. Mike Douglas will review the collaboration’s progress toward a physics of reasoning, organized around what is now measurable.
51:461.1K views · 4d ago · 51:46
Guest: Matthieu Wyart
Despite the fact that experimental neural scaling laws have substantially guided empirical progress in large-scale machine learning, no existing theory can quantitatively predict the exponents of these important laws for any modern LLM trained on natural languages. It is also unclear whether they can be beaten by training with other objectives.
48:111K views · 4d ago · 48:11
Guest: Eva Silverstein
How do modern neural networks evolve from randomly initialized weights toward a useful representation of the data? How does this proceed layerwise, and what role does the structure of attention play in the case of language models?
54:59629 views · 4d ago · 54:59
Guest: Cengiz Pehlevan
Training modern AI systems is increasingly expensive, yet many of the choices that determine this cost still depend on extensive empirical search. This talk will review how combining ideas from theoretical physics with targeted experiments make large-scale training more predictable, transferable, and efficient. Cengiz Pehlevan will begin by examining how parameterization and hyperparameter choices shape optimization dynamics.
59:05884 views · 4d ago · 59:05
Guest: Andrea Montanari
Classical statistical models are constructed so that either the number of parameters is small with respect to the sample size, or the regularization is strong enough so that their “effective dimension” is small. In contrast, modern machine learning models are vastly miss-specified and overparametrized, and only very loosely regularized.
Showing the 15 most recent episodes. Everything plays right here on The Pod Atlas.
Have thoughts about Simons Foundation? Share them with other listeners.
Write a review