Machine Teaching for Inverse Reinforcement Learning: Algorithms and Applications

TL;DR

This paper introduces a machine teaching framework for inverse reinforcement learning (IRL) to identify the minimal set of demonstrations needed to specify a reward function, improving efficiency and understanding of IRL processes.

Contribution

It formalizes the problem of teaching IRL with minimal demonstrations, reduces it to a set cover problem, and applies it to enhance IRL query efficiency and learning algorithms.

Findings

Developed an efficient approximation algorithm for maximally-informative demonstrations.

Provided a lower bound on the number of queries for active IRL.

Created a more efficient IRL algorithm using informative demonstrations.

Abstract

Inverse reinforcement learning (IRL) infers a reward function from demonstrations, allowing for policy improvement and generalization. However, despite much recent interest in IRL, little work has been done to understand the minimum set of demonstrations needed to teach a specific sequential decision-making task. We formalize the problem of finding maximally informative demonstrations for IRL as a machine teaching problem where the goal is to find the minimum number of demonstrations needed to specify the reward equivalence class of the demonstrator. We extend previous work on algorithmic teaching for sequential decision-making tasks by showing a reduction to the set cover problem which enables an efficient approximation algorithm for determining the set of maximally-informative demonstrations. We apply our proposed machine teaching algorithm to two novel applications: providing a lower bound on the number of queries needed to learn a policy using active IRL and developing a novel IRL algorithm that can learn more efficiently from informative demonstrations than a standard IRL approach.