Generating Adjacency-Constrained Subgoals in Hierarchical Reinforcement Learning

TL;DR

This paper introduces an adjacency constraint for hierarchical reinforcement learning that limits high-level goal generation to nearby states, improving training efficiency and performance in various control tasks.

Contribution

It proposes a novel adjacency constraint mechanism, theoretically preserves optimal policies, and demonstrates improved empirical results over existing HRL methods.

Findings

Enhanced training efficiency in HRL models

Improved performance on control benchmarks

Effective adjacency network implementation

Abstract

Goal-conditioned hierarchical reinforcement learning (HRL) is a promising approach for scaling up reinforcement learning (RL) techniques. However, it often suffers from training inefficiency as the action space of the high-level, i.e., the goal space, is often large. Searching in a large goal space poses difficulties for both high-level subgoal generation and low-level policy learning. In this paper, we show that this problem can be effectively alleviated by restricting the high-level action space from the whole goal space to a $k$-step adjacent region of the current state using an adjacency constraint. We theoretically prove that the proposed adjacency constraint preserves the optimal hierarchical policy in deterministic MDPs, and show that this constraint can be practically implemented by training an adjacency network that can discriminate between adjacent and non-adjacent subgoals. Experimental results on discrete and continuous control tasks show that incorporating the adjacency constraint improves the performance of state-of-the-art HRL approaches in both deterministic and stochastic environments.