Score: 0

Neural Policy Iteration for Stochastic Optimal Control: A Physics-Informed Approach

Published: August 3, 2025 | arXiv ID: 2508.01718v1

By: Yeongjong Kim , Yeoneung Kim , Minseok Kim and more

Potential Business Impact:

Helps robots learn tasks faster and more reliably.

We propose a physics-informed neural network policy iteration (PINN-PI) framework for solving stochastic optimal control problems governed by second-order Hamilton--Jacobi--Bellman (HJB) equations. At each iteration, a neural network is trained to approximate the value function by minimizing the residual of a linear PDE induced by a fixed policy. This linear structure enables systematic $L^2$ error control at each policy evaluation step, and allows us to derive explicit Lipschitz-type bounds that quantify how value gradient errors propagate to the policy updates. This interpretability provides a theoretical basis for evaluating policy quality during training. Our method extends recent deterministic PINN-based approaches to stochastic settings, inheriting the global exponential convergence guarantees of classical policy iteration under mild conditions. We demonstrate the effectiveness of our method on several benchmark problems, including stochastic cartpole, pendulum problems and high-dimensional linear quadratic regulation (LQR) problems in up to 10D.

Country of Origin
🇰🇷 Korea, Republic of

Page Count
9 pages

Category
Computer Science:
Machine Learning (CS)