Reward Learning using Structural Motifs in Inverse Reinforcement Learning

TL;DR

This paper introduces SMIRL, a novel IRL method that leverages structural motifs like finite-state automata to improve reward learning in complex, long-horizon tasks, outperforming existing methods in various environments.

Contribution

The paper proposes SMIRL, which incorporates task structure via automata into IRL, enhancing learning efficiency and success in complex tasks compared to baseline methods.

Findings

Successfully learns complex tasks where baselines fail

Outperforms baselines in sample efficiency on simpler tasks

Shows promising results in continuous domains with compositional rewards

Abstract

The Inverse Reinforcement Learning (\textit{IRL}) problem has seen rapid evolution in the past few years, with important applications in domains like robotics, cognition, and health. In this work, we explore the inefficacy of current IRL methods in learning an agent's reward function from expert trajectories depicting long-horizon, complex sequential tasks. We hypothesize that imbuing IRL models with structural motifs capturing underlying tasks can enable and enhance their performance. Subsequently, we propose a novel IRL method, SMIRL, that first learns the (approximate) structure of a task as a finite-state-automaton (FSA), then uses the structural motif to solve the IRL problem. We test our model on both discrete grid world and high-dimensional continuous domain environments. We empirically show that our proposed approach successfully learns all four complex tasks, where two foundational IRL baselines fail. Our model also outperforms the baselines in sample efficiency on a simpler toy task. We further show promising test results in a modified continuous domain on tasks with compositional reward functions.