Evidence of social learning across symbolic cultural barriers in sperm whales @SimonsInstituteTOC

Simons Institute | Evidence of social learning across symbolic cultural barriers in sperm whales @SimonsInstituteTOC | Uploaded 2 months ago | Updated 9 hours ago
Giovanni Petri (CentAI)
https://simons.berkeley.edu/talks/giovanni-petri-centai-2024-06-30
Decoding Communication in Nonhuman Species III (co-hosted with Project CETI)

Understanding (a bit about) hallucinations in Generative AI

Harnessing the properties of equivariant neural networks to understand and design materials

Some thoughts on ML-based protein engineering

Tractable Representations for Boolean Functional Synthesis

Generalizable sampling of conformational ensembles with latent space dynamics

Cut Sparsification and Succinct Representation of Submodular Hypergraphs

Graph Connectivity Using Star Contraction

$Almost-Optimal Sublinear Additive Spanners Zihan Tan (Rutgers University) https://simons.berkeley.edu/talks/zihan-tan-rutgers-university-2024-07-29 Sublinear Graph Simplification Given an undirected unweighted graph G=(V, E) on n vertices and m edges, a subgraph H is a spanner of G with stretch function f, iff for every pair s, t of vertices, dist_H(s,t) is at most f(dist_G(s,t)). When f(d)=d+o(d), H is called a sublinear additive spanner. In this talk, we show that for any constant integer k＞=2, every graph on n vertices has a sublinear additive spanner with stretch function f(d)=d+O(d^{1-1/k}) and O(n^{1+1/(2^{k+1}-1)+o(1)}) edges. The size bound of our spanners almost matches the lower bound proved in [Abboud-Bodwin-Pettie 2017], which holds for any data structure that maintains distances within the same stretch function. This talk is based on joint work with Tianyi Zhang$

Randomized Least Squares Optimization and its Incredible Utility for Large-Scale Tensor Decomp...