Multi-Task Reinforcement Learning with Soft Modularization

TL;DR

This paper introduces a soft modularization approach for multi-task reinforcement learning that dynamically reconfigures a base policy network for different tasks, improving efficiency and performance.

Contribution

It proposes a novel soft modularization technique with a routing network to better share parameters across tasks in reinforcement learning.

Findings

Improves sample efficiency in robotics manipulation tasks.

Achieves higher performance compared to strong baselines.

Effective in sequential multi-task learning scenarios.

Abstract

Multi-task learning is a very challenging problem in reinforcement learning. While training multiple tasks jointly allow the policies to share parameters across different tasks, the optimization problem becomes non-trivial: It remains unclear what parameters in the network should be reused across tasks, and how the gradients from different tasks may interfere with each other. Thus, instead of naively sharing parameters across tasks, we introduce an explicit modularization technique on policy representation to alleviate this optimization issue. Given a base policy network, we design a routing network which estimates different routing strategies to reconfigure the base network for each task. Instead of directly selecting routes for each task, our task-specific policy uses a method called soft modularization to softly combine all the possible routes, which makes it suitable for sequential tasks. We experiment with various robotics manipulation tasks in simulation and show our method improves both sample efficiency and performance over strong baselines by a large margin.