Efficient Reinforcement Learning Through Adaptively Pretrained Visual Encoder

TL;DR

This paper introduces APE, an adaptive pretraining framework for visual encoders in reinforcement learning, which enhances generalization and sampling efficiency across diverse tasks by using adaptive augmentation during pretraining.

Contribution

The paper presents a novel adaptive pretraining approach for visual encoders that improves generalization and efficiency in RL, outperforming existing methods across multiple benchmarks.

Findings

State-of-the-art performance with DreamerV3 and DrQ-v2 using APE.

Significant improvement in sampling efficiency with visual inputs.

Effective generalization across various domains.

Abstract

While Reinforcement Learning (RL) agents can successfully learn to handle complex tasks, effectively generalizing acquired skills to unfamiliar settings remains a challenge. One of the reasons behind this is the visual encoders used are task-dependent, preventing effective feature extraction in different settings. To address this issue, recent studies have tried to pretrain encoders with diverse visual inputs in order to improve their performance. However, they rely on existing pretrained encoders without further exploring the impact of pretraining period. In this work, we propose APE: efficient reinforcement learning through Adaptively Pretrained visual Encoder -- a framework that utilizes adaptive augmentation strategy during the pretraining phase and extracts generalizable features with only a few interactions within the task environments in the policy learning period. Experiments are conducted across various domains, including DeepMind Control Suite, Atari Games and Memory Maze benchmarks, to verify the effectiveness of our method. Results show that mainstream RL methods, such as DreamerV3 and DrQ-v2, achieve state-of-the-art performance when equipped with APE. In addition, APE significantly improves the sampling efficiency using only visual inputs during learning, approaching the efficiency of state-based method in several control tasks. These findings demonstrate the potential of adaptive pretraining of encoder in enhancing the generalization ability and efficiency of visual RL algorithms.