CIM: Constrained Intrinsic Motivation for Sparse-Reward Continuous Control

TL;DR

This paper introduces CIM, a novel approach that uses constrained intrinsic motivation and Lagrangian methods to improve exploration and sample efficiency in sparse-reward continuous control tasks.

Contribution

The paper proposes CIM, a new method that leverages task priors and adaptive balancing to enhance exploration in reinforcement learning with sparse rewards.

Findings

CIM outperforms state-of-the-art methods on multiple tasks.

CIM improves sample efficiency significantly.

Key techniques can enhance existing algorithms.

Abstract

Intrinsic motivation is a promising exploration technique for solving reinforcement learning tasks with sparse or absent extrinsic rewards. There exist two technical challenges in implementing intrinsic motivation: 1) how to design a proper intrinsic objective to facilitate efficient exploration; and 2) how to combine the intrinsic objective with the extrinsic objective to help find better solutions. In the current literature, the intrinsic objectives are all designed in a task-agnostic manner and combined with the extrinsic objective via simple addition (or used by itself for reward-free pre-training). In this work, we show that these designs would fail in typical sparse-reward continuous control tasks. To address the problem, we propose Constrained Intrinsic Motivation (CIM) to leverage readily attainable task priors to construct a constrained intrinsic objective, and at the same time, exploit the Lagrangian method to adaptively balance the intrinsic and extrinsic objectives via a simultaneous-maximization framework. We empirically show, on multiple sparse-reward continuous control tasks, that our CIM approach achieves greatly improved performance and sample efficiency over state-of-the-art methods. Moreover, the key techniques of our CIM can also be plugged into existing methods to boost their performances.