Factored Adaptation for Non-Stationary Reinforcement Learning

TL;DR

This paper introduces FANS-RL, a novel approach that models non-stationarity in reinforcement learning using causal structures and factored representations, improving adaptability and robustness in changing environments.

Contribution

It proposes a causal, factored adaptation framework for non-stationary RL that can recover underlying causal graphs and handle diverse non-stationary scenarios.

Findings

FANS-RL outperforms existing methods in return and robustness.

It achieves a compact latent state representation.

The approach effectively handles different types and frequencies of non-stationarity.

Abstract

Dealing with non-stationarity in environments (e.g., in the transition dynamics) and objectives (e.g., in the reward functions) is a challenging problem that is crucial in real-world applications of reinforcement learning (RL). While most current approaches model the changes as a single shared embedding vector, we leverage insights from the recent causality literature to model non-stationarity in terms of individual latent change factors, and causal graphs across different environments. In particular, we propose Factored Adaptation for Non-Stationary RL (FANS-RL), a factored adaption approach that learns jointly both the causal structure in terms of a factored MDP, and a factored representation of the individual time-varying change factors. We prove that under standard assumptions, we can completely recover the causal graph representing the factored transition and reward function, as well as a partial structure between the individual change factors and the state components. Through our general framework, we can consider general non-stationary scenarios with different function types and changing frequency, including changes across episodes and within episodes. Experimental results demonstrate that FANS-RL outperforms existing approaches in terms of return, compactness of the latent state representation, and robustness to varying degrees of non-stationarity.