Optimal-PhiBE: A PDE-based Model-free framework for Continuous-time Reinforcement Learning

TL;DR

This paper introduces Optimal-PhiBE, a PDE-based framework for continuous-time reinforcement learning that reduces discretization errors and sensitivity issues, enabling more accurate and model-free policy learning from discrete data.

Contribution

Optimal-PhiBE integrates discrete-time information into a continuous PDE framework, improving accuracy and robustness over existing methods in CTRL.

Findings

Optimal-PhiBE recovers the optimal policy exactly in the undiscounted case.

It outperforms Optimal-BE in weakly discounted or control-dominant scenarios.

Numerical experiments verify the theoretical error bounds and effectiveness.

Abstract

This paper addresses continuous-time reinforcement learning (CTRL) where the system dynamics are governed by an unknown stochastic differential equation, and only discrete-time observations are available. Existing approaches face limitations: model-based PDE methods suffer from non-identifiability, while model-free methods based on the discrete-time optimal Bellman equation (Optimal-BE) suffer from large discretization errors that are highly sensitive to both the system dynamics and the reward structure. To overcome these challenges, we introduce Optimal-PhiBE, a formulation that integrates discrete-time information into a continuous-time PDE, combining the strength of both existing frameworks while mitigating their limitations. Optimal-PhiBE exhibits smaller discretization errors when the uncontrolled system evolves slowly, and demonstrates reduced sensitivity to oscillatory reward structures, and enables model-free algorithms that bypass explicit dynamics estimation. In the linear-quadratic regulator (LQR) setting, sharp error bounds are established for both Optimal-PhiBE and Optimal-BE. The results show that Optimal-PhiBE exactly recovers the optimal policy in the undiscounted case and substantially outperforms Optimal-BE when the problem is weakly discounted or control-dominant. Furthermore, we extend Optimal-PhiBE to higher orders, providing increasingly accurate approximations. A model-free policy iteration algorithm is proposed to solve the Optimal-PhiBE directly from trajectory data. Numerical experiments are conducted to verify the theoretical findings.