Uploaded February 2026 | Updated September 2026, 2 weeks ago
Thomas Bourgeron (Institute Pasteur)
https://simons.berkeley.edu/talks/thomas-bourgeron-institute-pasteur-2026-02-10
Theory of Computing and Healthcare
Our group identified the first genes associated with autism pointing at a key role of the synapse in this complex condition. These findings underscore the genetic diversity of autism while revealing shared biological mechanisms. The genetic architecture of autism involves a complex combination of de novo, rare and common genetic variants shared with other traits such as attention deficit hyperactivity disorders (ADHD), cognitive skills, and epilepsy. In this presentation, I will introduce recent results that shed new light on different subgroups of people with autism and how they differ at the clinical, brain imaging and genetic levels. I will also illustrate how we can explore synaptic genes such as SHANK2 and SHANK3 and their roles in cognition and social motivation. Finally, I will present how we are currently using participatory research to study Risk, Resilience and Developmental Diversity in Mental Health (The R2D2-MH project) to understand why some carriers of genetic variants seem to be protected from adverse symptoms while others have more difficulties to thrive in the society.
Thomas Bourgeron (Institute Pasteur)
https://simons.berkeley.edu/talks/thomas-bourgeron-institute-pasteur-2026-02-10
Theory of Computing and Healthcare
Our group identified the first genes associated with autism pointing at a key role of the synapse in this complex condition. These findings underscore the genetic diversity of autism while revealing shared biological mechanisms. The genetic architecture of autism involves a complex combination of de novo, rare and common genetic variants shared with other traits such as attention deficit hyperactivity disorders (ADHD), cognitive skills, and epilepsy. In this presentation, I will introduce recent results that shed new light on different subgroups of people with autism and how they differ at the clinical, brain imaging and genetic levels. I will also illustrate how we can explore synaptic genes such as SHANK2 and SHANK3 and their roles in cognition and social motivation. Finally, I will present how we are currently using participatory research to study Risk, Resilience and Developmental Diversity in Mental Health (The R2D2-MH project) to understand why some carriers of genetic variants seem to be protected from adverse symptoms while others have more difficulties to thrive in the society.







![Latent Variable models and Subset Smoothing
Ravi Kannan (Simons Institute, UC Berkeley)
https://simons.berkeley.edu/talks/ravi-kannan-2026-05-26
The Role of TCS in Modern Machine Learning
A number of Latent Variable Models in Machine Learning (including Mixture Models, Topic Models, Stochastic block models and Mixed Membership Community Mod els) can be abstracted to the geometric problem of learn ing a latent polytope K given data points, each obtained by randomly perturbing a latent point in K. The challenge is that perturbations are typically much larger than the dimensions of K and so data points lie (far) outside K. To tackle this, we introduce the “Subset Smoothed” polytope K′ which is the convex hull of (n/k) points, each obtained by averaging a k− subset of the n data points. [k is a parameter.] We will observe that K′ ≈ K under reasonable assumptions on data. We will also observe that K′ has a polynomial time optimization oracle. These simple observations are the starting point of our provable algorithm for learning K which the talk will describe.
Joint Work with Chiranjib Bhattacharyya, Amit Kumar Latent Variable models and Subset Smoothing](https://i.ytimg.com/vi/Dm1YnND7Qmo/mqdefault.jpg)


