大阪大学理学研究科　フォアフロント研究センター　理論連携研究(TJR)プロジェクト

第２回 TJRセミナー https://www.phys.sci.osaka-u.ac.jp/nambu/tjr/

Marc Mezard教授は、スピングラスから構造ガラス、情報統計力学にまで分野横断的に広がったランダム系の統計力学の各分野において、最前線を切り開くマイルストーンとなる重要な研究をされてこられました。

今回は、生成AIでお馴染みの拡散モデルについての最近の統計力学的解析についてお話しくださいます。

いつも明快な講演をされます。ご期待ください。

問い合わせ先：吉野 元 yoshino.hajime.cmc@osaka-u.ac.jp

Abstract:

Generative models, in which one trains an algorithm to generate samples ‘similar’ to those of a data base, is a major new direction developed in machine learning in the recent years. In particular, generative models based on diffusion equations have become the state of the art, notably for image generation. However, the reasons for this spectacular technological success are not well understood, and neither are its limitations.

After an introduction to this topic, the talk will focus on the behavior of generative diffusion in the high-dimensional limit, where data are formed by a very large number of variables.  Using methods from statistical physics, and through a detailed analysis of two well-controlled high-dimensional cases, we explain the various phase transitions that take place during the dynamics of generation.

 Giulio Biroli, Tony Bonnaire, Valentin de Bortoli, Marc Mézard, "Dynamical Regimes of Diffusion Models",  arXiv:2402.18491v1

Giulio Biroli, Marc Mézard, "Generative diffusion in very large dimensions", J. Stat. Mech. (2023) 093402.