Timezone: »

 
Poster
Recurrent Hypernetworks are Surprisingly Strong in Meta-RL
Jacob Beck · Risto Vuorio · Zheng Xiong · Shimon Whiteson

Thu Dec 14 08:45 AM -- 10:45 AM (PST) @ Great Hall & Hall B1+B2 #1310
in Poster Session 5 »
Event URL: https://github.com/jacooba/hyper »

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.

Author Information

Jacob Beck (University of Oxford)
Risto Vuorio (University of Oxford)

I'm a PhD student in WhiRL at University of Oxford. I'm interested in reinforcement learning and meta-learning.

Zheng Xiong (University of Oxford)
Shimon Whiteson (Oxford University)

More from the Same Authors