Guided Cooperation in Hierarchical Reinforcement Learning via Model-based Rollout

TL;DR

This paper introduces GCMR, a goal-conditioned hierarchical reinforcement learning framework that enhances inter-level cooperation and stability through model-based rollout, gradient penalties, and guided planning, leading to improved policy performance.

Contribution

The paper proposes GCMR, a novel HRL framework that integrates model-based rollout, gradient constraints, and one-step planning to improve inter-level cooperation and exploration.

Findings

GCMR improves sample efficiency and policy stability.

GCMR outperforms state-of-the-art algorithms in experiments.

Incorporating GCMR yields more robust hierarchical policies.

Abstract

Goal-conditioned hierarchical reinforcement learning (HRL) presents a promising approach for enabling effective exploration in complex, long-horizon reinforcement learning (RL) tasks through temporal abstraction. Empirically, heightened inter-level communication and coordination can induce more stable and robust policy improvement in hierarchical systems. Yet, most existing goal-conditioned HRL algorithms have primarily focused on the subgoal discovery, neglecting inter-level cooperation. Here, we propose a goal-conditioned HRL framework named Guided Cooperation via Model-based Rollout (GCMR), aiming to bridge inter-layer information synchronization and cooperation by exploiting forward dynamics. Firstly, the GCMR mitigates the state-transition error within off-policy correction via model-based rollout, thereby enhancing sample efficiency. Secondly, to prevent disruption by the unseen subgoals and states, lower-level Q-function gradients are constrained using a gradient penalty with a model-inferred upper bound, leading to a more stable behavioral policy conducive to effective exploration. Thirdly, we propose a one-step rollout-based planning, using higher-level critics to guide the lower-level policy. Specifically, we estimate the value of future states of the lower-level policy using the higher-level critic function, thereby transmitting global task information downwards to avoid local pitfalls. These three critical components in GCMR are expected to facilitate inter-level cooperation significantly. Experimental results demonstrate that incorporating the proposed GCMR framework with a disentangled variant of HIGL, namely ACLG, yields more stable and robust policy improvement compared to various baselines and significantly outperforms previous state-of-the-art algorithms.