Join us on Wednesday, November 13th, at 12:30 PM EDT for a talk by Alireza Mousavi (UofT) on "Learning and Optimization with Mean-Field Langevin Dynamics" at Mila - Institut québécois d'IA in Montreal

Honored to receive this fellowship. Thank you RBC Borealis for supporting our research, and for the cool sweatshirts :)

So can someone ask o3 to prove that with high probability over initialization, gradient descent on ResNet and CIFAR10 converges to 0 loss?

Congratulations to Vector-affiliated researchers Alireza Mousavi-Hosseini and Mohammed Adnan, who were named RBC Borealis Fellows. The RBC Borealis Fellowships Program represents excellence in Canadian AI research and innovation. We’re proud to see our affiliated researchers

Excited to announce a new paper with Yunwei Ren, Denny Wu, Jason Lee! We prove a neural scaling law in the SGD learning of extensive width two-layer neural networks. arxiv.org/abs/2504.19983 🧵below (1/10)

New work arxiv.org/abs/2506.05500 on learning multi-index models with Alex Damian and Joan Bruna. Multi-index are of the form y= g(Ux), where U=r by d maps from d dimension to r dimension and d>>r. g is an arbitrary function. Examples of multi-index models are any neural net

Why does RL struggle with tasks requiring long reasoning chains? Because “bumping into” a correct solution becomes exponentially less likely as the number of reasoning steps grows. We propose an adaptive backtracking algorithm: AdaBack. 1/n

NeurIPS Conference, why take the option to provide figures in the rebuttals away from the authors during the rebuttal period? Grounding the discussion in hard evidential data (like plots) makes resolving disagreements much easier for both the authors and the reviewers. Left: NeurIPS