Learning Time-Invariant Reward Functions through Model-Based Inverse Reinforcement Learning

TL;DR

This paper introduces a method for learning time-invariant reward functions in inverse reinforcement learning, enabling generalization across different execution durations and relaxed temporal alignment, demonstrated on robotic tasks.

Contribution

The authors propose a novel formulation for learning time-invariant costs that relaxes temporal alignment constraints in inverse reinforcement learning.

Findings

Enables learning temporally invariant rewards from misaligned demonstrations.

Generalizes spatially to out-of-distribution tasks.

Effective on simulated robotic placement and peg-in-hole tasks.

Abstract

Inverse reinforcement learning is a paradigm motivated by the goal of learning general reward functions from demonstrated behaviours. Yet the notion of generality for learnt costs is often evaluated in terms of robustness to various spatial perturbations only, assuming deployment at fixed speeds of execution. However, this is impractical in the context of robotics and building, time-invariant solutions is of crucial importance. In this work, we propose a formulation that allows us to 1) vary the length of execution by learning time-invariant costs, and 2) relax the temporal alignment requirements for learning from demonstration. We apply our method to two different types of cost formulations and evaluate their performance in the context of learning reward functions for simulated placement and peg in hole tasks executed on a 7DoF Kuka IIWA arm. Our results show that our approach enables learning temporally invariant rewards from misaligned demonstration that can also generalise spatially to out of distribution tasks.