Teachable Reinforcement Learning via Advice Distillation

TL;DR

This paper introduces a new interactive learning paradigm where agents learn from structured advice provided by a teacher, reducing the need for extensive human supervision compared to traditional reinforcement and imitation learning methods.

Contribution

It formalizes a class of human-in-the-loop decision problems and proposes a simple algorithm enabling agents to interpret and learn from advice, improving efficiency in skill acquisition.

Findings

Agents learn new skills with less human supervision.

Advice-based learning outperforms standard reinforcement learning.

Agents often require less supervision than imitation learning.

Abstract

Training automated agents to complete complex tasks in interactive environments is challenging: reinforcement learning requires careful hand-engineering of reward functions, imitation learning requires specialized infrastructure and access to a human expert, and learning from intermediate forms of supervision (like binary preferences) is time-consuming and extracts little information from each human intervention. Can we overcome these challenges by building agents that learn from rich, interactive feedback instead? We propose a new supervision paradigm for interactive learning based on "teachable" decision-making systems that learn from structured advice provided by an external teacher. We begin by formalizing a class of human-in-the-loop decision making problems in which multiple forms of teacher-provided advice are available to a learner. We then describe a simple learning algorithm for these problems that first learns to interpret advice, then learns from advice to complete tasks even in the absence of human supervision. In puzzle-solving, navigation, and locomotion domains, we show that agents that learn from advice can acquire new skills with significantly less human supervision than standard reinforcement learning algorithms and often less than imitation learning.