A Fisher-Rao Gradient Flow for Entropic Mean-Field Min-Max Games

Abstract

Gradient flows play a substantial role in addressing many machine learning problems. We examine the convergence in continuous-time of a Fisher-Rao (Mean-Field Birth-Death) gradient flow in the context of solving convex-concave min-max games with entropy regularization. We propose appropriate Lyapunov functions to demonstrate convergence with explicit rates to the unique mixed Nash equilibrium.

Cite

Text

Lascu et al. "A Fisher-Rao Gradient Flow for Entropic Mean-Field Min-Max Games." Transactions on Machine Learning Research, 2024.

Markdown

[Lascu et al. "A Fisher-Rao Gradient Flow for Entropic Mean-Field Min-Max Games." Transactions on Machine Learning Research, 2024.](https://mlanthology.org/tmlr/2024/lascu2024tmlr-fisherrao/)

BibTeX

@article{lascu2024tmlr-fisherrao,
  title     = {{A Fisher-Rao Gradient Flow for Entropic Mean-Field Min-Max Games}},
  author    = {Lascu, Razvan-Andrei and Majka, Mateusz B. and Szpruch, Lukasz},
  journal   = {Transactions on Machine Learning Research},
  year      = {2024},
  url       = {https://mlanthology.org/tmlr/2024/lascu2024tmlr-fisherrao/}
}