Fast Task-Adaptation for Tasks Labeled Using Natural Language in Reinforcement Learning

TL;DR

This paper explores how natural language instructions can facilitate rapid and effective adaptation of reinforcement learning policies to new tasks, enabling lifelong learning and better generalization.

Contribution

It introduces a method to evaluate and select the best base control policy for new tasks using natural language instructions, improving adaptation efficiency.

Findings

Method effectively identifies suitable policies for unseen tasks

Natural language instructions enhance adaptation speed

Supports lifelong learning with policy selection

Abstract

Over its lifetime, a reinforcement learning agent is often tasked with different tasks. How to efficiently adapt a previously learned control policy from one task to another, remains an open research question. In this paper, we investigate how instructions formulated in natural language can enable faster and more effective task adaptation. This can serve as the basis for developing language instructed skills, which can be used in a lifelong learning setting. Our method is capable of assessing, given a set of developed base control policies, which policy will adapt best to a new unseen task.