3 papers
cs.LG2026
Understanding diffusion models requires rethinking (again) generalization
Pierre Marion, Yu-Han Wu
This position paper argues that understanding generalization in diffusion models requires fundamentally new theoretical frameworks that go beyond both classical statistical learnin…
cs.LG2025
Phase Diagram of Dropout for Two-Layer Neural Networks in the Mean-Field Regime
Lénaïc Chizat, Pierre Marion, Yerkin Yesbay
Dropout is a standard training technique for neural networks that consists of randomly deactivating units at each step of their gradient-based training. It is known to improve perf…
stat.ML2024
Deep linear networks for regression are implicitly regularized towards flat minima
Pierre Marion, Lénaïc Chizat
The largest eigenvalue of the Hessian, or sharpness, of neural networks is a key quantity to understand their optimization dynamics. In this paper, we study the sharpness of deep l…