Action-Free Offline-to-Online RL via Discretised State Policies

TL;DR

This paper introduces a new offline-to-online reinforcement learning framework that learns from datasets without actions by focusing on state transitions, using discretisation and pre-trained state policies to improve online learning efficiency.

Contribution

It proposes a novel state discretisation transformation and a value-based algorithm, Offline State-Only DecQN, for pre-training from action-free data, and introduces guided online learning leveraging these policies.

Findings

Improves convergence speed in online RL tasks.

Enhances asymptotic performance with action-free pre-training.

Discretisation and regularisation are key to effectiveness.

Abstract

Most existing offline RL methods presume the availability of action labels within the dataset, but in many practical scenarios, actions may be missing due to privacy, storage, or sensor limitations. We formalise the setting of action-free offline-to-online RL, where agents must learn from datasets consisting solely of $(s,r,s')$ tuples and later leverage this knowledge during online interaction. To address this challenge, we propose learning state policies that recommend desirable next-state transitions rather than actions. Our contributions are twofold. First, we introduce a simple yet novel state discretisation transformation and propose Offline State-Only DecQN (\algo), a value-based algorithm designed to pre-train state policies from action-free data. \algo{} integrates the transformation to scale efficiently to high-dimensional problems while avoiding instability and overfitting associated with continuous state prediction. Second, we propose a novel mechanism for guided online learning that leverages these pre-trained state policies to accelerate the learning of online agents. Together, these components establish a scalable and practical framework for leveraging action-free datasets to accelerate online RL. Empirical results across diverse benchmarks demonstrate that our approach improves convergence speed and asymptotic performance, while analyses reveal that discretisation and regularisation are critical to its effectiveness.