Workshop II: PDE and Inverse Problem Methods in Machine Learning

Part of the Long Program High Dimensional Hamilton-Jacobi PDEs

April 20 - 24, 2020

Poster Session

Welcome to the Workshop II: “PDE and Inverse Problem Methods in Machine Learning” (HJWS2) Virtual Poster Session

Registered participants will receive an email to participate in the online poster session – to be held between April 20-24, 2020 – to view and comment on the virtual posters. Presenters will be available during that week to respond to comments on their posters. After April 24th, all virtual posters will remain publicly available as an archive.

Poster #1: “A Mean-field Analysis of Deep ResNet and Beyond: Towards Provable Optimization Via Overparameterization From Depth” Yiping Lu (Stanford)

Poster #2: “You Only Propagate Once: Painless Adversarial Training Using Maximal Principle” Yiping Lu (Stanford)

Poster #3: “Representing exact viscosity solutions to high dimensional Hamilton–Jacobi PDEs using neural network architectures” Tingwei Meng (Brown)

Poster #4: “Differential Equation Recovery Using Machine Learning” Elisa Negrini (Worcester Polytechnic Univ.)

Poster #5: “Krylov subspace type methods for the computation of non-negative or sparse solutions of ill-posed problems“ Mirjeta Pasha (Kent State)

Poster #6: “Solving high-dimensional Hamilton-Jacobi PDEs using neural networks: perspectives from the theory of controlled diffusions and measures on path space” Lorenz Richter (FU Berlin)