Physics-Informed Representation and Learning: Control and Risk Quantification

TL;DR

This paper introduces a physics-informed neural network framework combined with dimensionality reduction for efficient optimal and safe control of high-dimensional stochastic systems, validated through experiments.

Contribution

It develops a novel approach integrating physics-informed neural networks with autoencoder-based feature extraction for control and risk quantification in complex systems.

Findings

Substantial sample efficiency improvement over existing methods

Effective low-dimensional feature identification via autoencoder-like neural network

Validated through experiments and quantitative analysis

Abstract

Optimal and safety-critical control are fundamental problems for stochastic systems, and are widely considered in real-world scenarios such as robotic manipulation and autonomous driving. In this paper, we consider the problem of efficiently finding optimal and safe control for high-dimensional systems. Specifically, we propose to use dimensionality reduction techniques from a comparison theorem for stochastic differential equations together with a generalizable physics-informed neural network to estimate the optimal value function and the safety probability of the system. The proposed framework results in substantial sample efficiency improvement compared to existing methods. We further develop an autoencoder-like neural network to automatically identify the low-dimensional features of the system to enhance the ease of design for system integration. We also provide experiments and quantitative analysis to validate the efficacy of the proposed method. Source code is available at https://github.com/jacobwang925/path-integral-PINN.