Meet GUD: our new Generative Unified Diffusion model inspired by the Renormalization Group. It generalizes standard diffusion by a choice of basis, a prior and separate noise schedules per dimension. Main author Mathis Gerdes and w/ Miranda Cheng. arxiv.org/abs/2410.02667

Microsoft launched the best course on Generative AI! The free 18 lesson course is available on Github and will teach you everything you need to know to start building Generative AI applications.

Today’s recommendation

Congratulations Dr Antonin Schrab on passing the PhD defence with no corrections!! Thanks to Rajen Shah and Carlo Ciliberto for their superb thesis examining work!

Many congratulations to David Baker, Gatsby Unit alumnus Demis Hassabis (Demis Hassabis), and John Jumper!

A new concentration bound on the smallest eigenvalue of gram matrix covering both low and high dimension cases! arxiv.org/abs/2410.05754

I moved from Toronto to the London with Geoffrey Hinton as one of the founding faculty of the Gatsby Computational Neuroscience Unit Unit. A few years afterwards, Demis Hassabis did his PostDoc there. Congrats to the Unit; what an amazing place! amp.theguardian.com/science/2024/o…

UC Berkeley's "Machine Learning" lecture notes people.eecs.berkeley.edu/~jrs/papers/ma…

Work from our amazing collaborator where they study theoretical foundation of CoT! arxiv.org/abs/2410.08633

I will be speaking about our recent series of papers on learning in infinite dimensions! Joint Seminar on Inverse Problems and Learning Theory on Oct 29, 14:00 CET (online). Seminar homepage: sites.google.com/view/joint-sem…

Only about 800 papers on diffusion models submitted to ICLR. Will read them over the weekend.

Thrilled to share that I’ll be joining Google DeepMind as a Research Scientist with the Discovery Team! It’s a dream come true—8 years ago, I watched AlphaGo live in high school, and now I am lucky enough to be part of this incredible journey. Can’t wait to discover what’s next!

After an incredible journey, I’ve finished my study as PhD student at UCL and am stepping into a new chapter as an Assistant Professor at the University of Bristol. From an uncertain undergrad to overcoming countless challenges along the way of research, there were many days

Conditional mean embeddings without Tikhonov: no saturation effect! "Optimal Rates for Vector-Valued Spectral Regularization Learning Algorithms" #NeurIPS2024 Poster #5709 12 Dec 11 a.m.— 2 p.m neurips.cc/virtual/2024/p… Dimitri Meunier ,Shen,Mattes Mollenhauer , Michael Li

✈️ to Nice😁 I’ll be giving an invited talk at IMS-ICSDS this year! Let me know if you are here and let’s discuss🫡

Here is everything that happened in AI Agents this week 🧵  (save for later)

ICLR2025に7本の論文が採択されました😃 We got 7 papers accepted by ICLR2025!

TL;DR: ✅ Theoretical guarantees for nonlinear meta-learning ✅ Explains when and how aggregation helps ✅ Connects RKHS regression, subspace estimation & meta-learning Co-led with Michael Li 🙌, with invaluable support from Arthur Gretton, Samory Kpotufe.