Long-Term Visitation Value for Deep Exploration in Sparse Reward Reinforcement Learning

TL;DR

This paper introduces a novel model-free, off-policy reinforcement learning method that uses long-term visitation counts to improve exploration in environments with sparse rewards, outperforming existing methods especially with suboptimal reward modes.

Contribution

It proposes a new exploration strategy based on long-term visitation values and decouples exploration from exploitation, along with new benchmarks for evaluation.

Findings

Outperforms existing exploration methods in sparse reward environments

Scales well with environment size

Effective in environments with suboptimal reward modes

Abstract

Reinforcement learning with sparse rewards is still an open challenge. Classic methods rely on getting feedback via extrinsic rewards to train the agent, and in situations where this occurs very rarely the agent learns slowly or cannot learn at all. Similarly, if the agent receives also rewards that create suboptimal modes of the objective function, it will likely prematurely stop exploring. More recent methods add auxiliary intrinsic rewards to encourage exploration. However, auxiliary rewards lead to a non-stationary target for the Q-function. In this paper, we present a novel approach that (1) plans exploration actions far into the future by using a long-term visitation count, and (2) decouples exploration and exploitation by learning a separate function assessing the exploration value of the actions. Contrary to existing methods which use models of reward and dynamics, our approach is off-policy and model-free. We further propose new tabular environments for benchmarking exploration in reinforcement learning. Empirical results on classic and novel benchmarks show that the proposed approach outperforms existing methods in environments with sparse rewards, especially in the presence of rewards that create suboptimal modes of the objective function. Results also suggest that our approach scales gracefully with the size of the environment. Source code is available at https://github.com/sparisi/visit-value-explore