Solving Inverse Physics Problems with Score Matching

Benjamin Holzschuh, Simona Vegetti, Nils Thuerey

Keywords: inverse problems, diffusion models, learned corrections, score matching

TL;DR: We propose a novel methodology for solving inverse problems that involve the temporal evolution of physical systems.

Abstract:

We propose to solve inverse problems involving the temporal evolution of physics systems by leveraging recent advances from diffusion models.
Our method moves the system's current state backward in time step by step by combining an approximate inverse physics simulator and a learned correction function.
A central insight of our work is that training the learned correction with a single-step loss is equivalent to a score matching objective, while recursively predicting longer parts of the trajectory during training relates to maximum likelihood training of a corresponding probability flow.
We highlight the advantages of our algorithm compared to standard denoising score matching and implicit score matching, as well as fully learned baselines for a wide range of inverse physics problems. The resulting inverse solver has excellent accuracy and temporal stability and, in contrast to other learned inverse solvers, allows for sampling the posterior of the solutions. Code and experiments are available at https://github.com/tum-pbs/SMDP.

Submission Number: 1882