Yi Ma

This account is verified. Learn more

@YiMaTweets

Chair Professor of AI, Hong Kong University
Pursuing a Mathematical Theory of Intelligence:
ma-lab-berkeley.github.io/deep-represent…

Hong Kong and Berkeley people.eecs.berkeley.edu/~yima/ Joined March 2022

845 Following 130K Followers 2.7K Posts

Latest posts 5

Pinned post

Yi Ma @YiMaTweets Oct 22

The lectures of my new course "Principles of Deep Representation Learning" are now available at the Youtube site: youtube.com/playlist?list=… with new lectures continuing to be added there. I will open-source all the slides later in the textbook website: ma-lab-berkeley.github.io/deep-represent…

www.youtube.com Principles of Deep Representation Learning | HKU DATA8014 Principles of Deep Representation Learning Fall 2026 · HKU DATA8014 Course Textbook (Open Access): https://ma-lab-berkeley.github.io/deep-representation-lear...

Yi Ma @YiMaTweets 12h

Congratulations to @EmmanuelCandes for receiving the Shaw Prize in mathematics for his pioneering work in compressed sensing and low-rank matrix recovery.

Yi Ma @YiMaTweets Oct 30

Here is the paper and code: arxiv.org/abs/2610.33337

arxiv.org Information-Theoretic Scaling Laws for Generative Diffusion Training Diffusion models trained on increasingly large datasets sampled from the same data distribution will converge to the same learned distribution. However, in practice, a model trained on finite data...

Dejan Stančević @DejanStancevic Oct 28

Check out our new work with @druv_pai, @YiMaTweets & @LucaAmb.

Paper: arxiv.org/abs/2610.33337

Code: github.com/DruvPai/diffus…

9/9
View quoted post on X

Yi Ma @YiMaTweets Oct 30

A theoretical sound approach to quantitatively characterize the generalization and memorization of diffusion models. Surprisingly elegant and computationally efficient and effective.

Luca Ambrogioni Verified account @LucaAmb Oct 28

Excited to share our work on information-theoretic scaling laws for diffusion models, with @druv_pai and @YiMaTweets!

By quantifying how much information a model learns from data, we can predict when memorization begins and how generation quality (FID) scales with training.
View quoted post on X

Yi Ma @YiMaTweets Oct 28

I am most interested in the scientific and computational principles behind our neural systems, especially the brain. I also believe, a systematic and rigorous scientific, mathematical, and computational theory of our neural systems (intelligence and consciousness) is the final frontier of Science.

Doris Tsao Verified account @doristsao Oct 27

"What we [in neurotech] do in the next ten years will determine how AI and humans co-exist" - Juan Benet.

I couldn't agree more. Juan and team @protocollabs are true visionaries pushing forward this incomparably important frontier.
View quoted post on X