Recurrent Hypernetworks are Surprisingly Strong in Meta-RL

TL;DR

This paper demonstrates that recurrent hypernetworks significantly improve meta-reinforcement learning performance, outperforming specialized methods by leveraging hypernetworks to enhance simple recurrent models.

Contribution

The study shows that combining hypernetworks with recurrent models yields surprisingly strong meta-RL performance, surpassing more complex specialized approaches.

Findings

Recurrent hypernetworks outperform existing meta-RL methods.

Hypernetworks are crucial for maximizing recurrent model performance.

Simple recurrent hypernetworks achieve state-of-the-art results.

Abstract

Deep reinforcement learning (RL) is notoriously impractical to deploy due to sample inefficiency. Meta-RL directly addresses this sample inefficiency by learning to perform few-shot learning when a distribution of related tasks is available for meta-training. While many specialized meta-RL methods have been proposed, recent work suggests that end-to-end learning in conjunction with an off-the-shelf sequential model, such as a recurrent network, is a surprisingly strong baseline. However, such claims have been controversial due to limited supporting evidence, particularly in the face of prior work establishing precisely the opposite. In this paper, we conduct an empirical investigation. While we likewise find that a recurrent network can achieve strong performance, we demonstrate that the use of hypernetworks is crucial to maximizing their potential. Surprisingly, when combined with hypernetworks, the recurrent baselines that are far simpler than existing specialized methods actually achieve the strongest performance of all methods evaluated. We provide code at https://github.com/jacooba/hyper.