no code implementations • 12 Feb 2023 • Julien Claisse, Giovanni Conforti, Zhenjie Ren, SongBo Wang
In this paper by adding the Fisher Information as the regularizer, we relate the regularized mean field optimization problem to a so-called mean field Schrodinger dynamics.
no code implementations • 6 Dec 2022 • Fan Chen, Zhenjie Ren, SongBo Wang
We study the mean field Langevin dynamics and the associated particle system.
no code implementations • 29 Jul 2020 • Anna Kazeykina, Zhenjie Ren, Xiaolu Tan, Junjian Yang
The results on the MFL equation can be applied to study the convergence of the Hamiltonian gradient descent algorithm for the overparametrized optimization.
no code implementations • 6 Apr 2020 • Giovanni Conforti, Anna Kazeykina, Zhenjie Ren
As applications, the dynamic games can be treated as games on a random environment when one treats the time horizon as the environment.
no code implementations • 16 Sep 2019 • Kaitong Hu, Anna Kazeykina, Zhenjie Ren
We introduce a system of mean-field Langevin equations, the invariant measure of which is shown to be the optimal control of the initial problem under mild conditions.
no code implementations • 19 May 2019 • Kaitong Hu, Zhenjie Ren, David Siska, Lukasz Szpruch
Our work is motivated by a desire to study the theoretical underpinning for the convergence of stochastic gradient type algorithms widely used for non-convex learning tasks such as training of neural networks.
