ODE-based Recurrent Model-free Reinforcement Learning for POMDPs

TL;DR

This paper introduces an ODE-based recurrent model within a model-free reinforcement learning framework to effectively infer unobservable information in POMDPs, demonstrating robustness and improved performance in continuous control and meta-RL tasks.

Contribution

It presents a novel ODE-based recurrent approach that enhances information inference in POMDPs within a model-free RL setting, especially under irregular observation sampling.

Findings

Effective in various PO continuous control tasks

Robust against irregularly-sampled observations

Outperforms baseline methods in POMDP scenarios

Abstract

Neural ordinary differential equations (ODEs) are widely recognized as the standard for modeling physical mechanisms, which help to perform approximate inference in unknown physical or biological environments. In partially observable (PO) environments, how to infer unseen information from raw observations puzzled the agents. By using a recurrent policy with a compact context, context-based reinforcement learning provides a flexible way to extract unobservable information from historical transitions. To help the agent extract more dynamics-related information, we present a novel ODE-based recurrent model combines with model-free reinforcement learning (RL) framework to solve partially observable Markov decision processes (POMDPs). We experimentally demonstrate the efficacy of our methods across various PO continuous control and meta-RL tasks. Furthermore, our experiments illustrate that our method is robust against irregular observations, owing to the ability of ODEs to model irregularly-sampled time series.