When we derived SignSGD and Signum, it was surprising how just keeping the sign of stochastic gradients is sufficient for convergence and we could prove that theoretically. In practice, Signum and derivatives like Lion work well even at large scale. It is interesting to see Adam
@animaanandkumar
-
FunDPS: Guided Diffusion Sampling Framework for PDE Solutions
By
–
Excited to introduce our latest work, Guided Diffusion Sampling on Function Spaces (FunDPS) (
https://
arxiv.org/abs/2505.17004) – a discretization-agnostic generative framework for solving PDE-based forward and inverse problems. Diffusion-based posterior sampling on function spaces: Our -
AI Understanding Physical World for Scientific Discovery
By
–
It was an honor to be part of @Google IO Dialogues stage with James Manyika, Pushmeet Kohli and Joëlle Barral and talk about AI+Science. I talked about how AI needs to understand the physical world in order to make new scientific discoveries. While LLMs can come up with new
-
Beyond LLMs: AI’s Role in Scientific Discovery
By
–
In a recent interview I talk about what it takes for AI to make new scientific discoveries. tldr: it won’t be just LLMs.
-
AI and Science Panel at Google IO 2026
By
–
Great to be part of the panel on ai+science at @Google IO
-
LeanDojo: Open-Source Framework for AI Theorem Proving
By
–
LeanDojo was the first opensource LLM+Lean framework. We continue to build and have new tools like Lean Copilot, Lean Agent, and Lean Progress that significantly enhance theorem proving workflows for mathematicians. All of our code is here : http://
github.com/lean-dojo/ Lean Dojo -
LeanDojo: Open Source LLM Framework Enhances Theorem Proving
By
–
LeanDojo was the first opensource LLM+Lean framework. We continue to build and have new tools like Lean Copilot, Lean Agent, and Lean Progress that significantly enhance theorem proving workflows for mathematicians. All of our code is here : http://
github.com/lean-dojo/ Lean Dojo -
Dictionary Learning Methods Show Promise for Data Representation
By
–
Cool to see dictionary learning come back. We had a work from decade ago showing nice properties for local methods to recover good dictionaries
-
Neural Operators for Learning PDEs Survey
By
–
We proposed neural operators for learning PDEs. You can see a survey here https://
rdcu.be/dD6LO -
AI as Scientific Tool: Weather Modeling and Discovery
By
–
Thank you @eostudi0 for coming to @Caltech and interviewing me on #ai I talk about the need to keep being curious and use AI as a tool, rather than being afraid of AI. I talk about AI for scientific modeling and discovery, and training the first high-resolution AI-based weather