Recomposing the Reinforcement Learning Building Blocks with Hypernetworks

TL;DR

This paper introduces a hypernetwork-based architecture for reinforcement learning building blocks, enhancing gradient estimation and reducing variance, leading to faster learning and better performance across various tasks and algorithms.

Contribution

It proposes a novel hypernetwork approach that explicitly models interactions between input variables in RL and Meta-RL, improving upon standard concatenation methods.

Findings

Improved gradient approximation in actor-critic algorithms.

Reduced variance in learning steps in Meta-RL.

Consistent performance gains across multiple tasks and algorithms.

Abstract

The Reinforcement Learning (RL) building blocks, i.e. Q-functions and policy networks, usually take elements from the cartesian product of two domains as input. In particular, the input of the Q-function is both the state and the action, and in multi-task problems (Meta-RL) the policy can take a state and a context. Standard architectures tend to ignore these variables' underlying interpretations and simply concatenate their features into a single vector. In this work, we argue that this choice may lead to poor gradient estimation in actor-critic algorithms and high variance learning steps in Meta-RL algorithms. To consider the interaction between the input variables, we suggest using a Hypernetwork architecture where a primary network determines the weights of a conditional dynamic network. We show that this approach improves the gradient approximation and reduces the learning step variance, which both accelerates learning and improves the final performance. We demonstrate a consistent improvement across different locomotion tasks and different algorithms both in RL (TD3 and SAC) and in Meta-RL (MAML and PEARL).