Hierarchical and Interpretable Skill Acquisition in Multi-task Reinforcement Learning

TL;DR

This paper introduces a hierarchical, interpretable multi-task reinforcement learning framework that enables agents to learn, reuse, and acquire new skills efficiently, with human-readable descriptions guiding skill selection.

Contribution

It proposes a novel hierarchical policy framework with temporal grammar for interpretable multi-task RL, allowing continual skill acquisition and reuse based on human language descriptions.

Findings

Agents effectively reuse skills in Minecraft tasks.

The framework improves learning efficiency for complex multi-task environments.

Agents provide human-interpretable descriptions of their decisions.

Abstract

Learning policies for complex tasks that require multiple different skills is a major challenge in reinforcement learning (RL). It is also a requirement for its deployment in real-world scenarios. This paper proposes a novel framework for efficient multi-task reinforcement learning. Our framework trains agents to employ hierarchical policies that decide when to use a previously learned policy and when to learn a new skill. This enables agents to continually acquire new skills during different stages of training. Each learned task corresponds to a human language description. Because agents can only access previously learned skills through these descriptions, the agent can always provide a human-interpretable description of its choices. In order to help the agent learn the complex temporal dependencies necessary for the hierarchical policy, we provide it with a stochastic temporal grammar that modulates when to rely on previously learned skills and when to execute new skills. We validate our approach on Minecraft games designed to explicitly test the ability to reuse previously learned skills while simultaneously learning new skills.