Model Predictive Adversarial Imitation Learning for Planning from Observation
Abstract
Human demonstration data is often ambiguous and incomplete, motivating imitation learning approaches that also exhibit reliable planning behavior. A common paradigm to perform planning-from-demonstration involves learning a reward function via Inverse Reinforcement Learning (IRL) then deploying this reward via Model Predictive Control (MPC). Towards unifying these methods, we derive a replacement of the policy in IRL with a planning-based agent. With connections to Adversarial Imitation Learning, this formulation enables end-to-end interactive learning of planners from observation-only demonstrations. In addition to benefits in interpretability, complexity, and safety, we study and observe significant improvements on sample efficiency, out-of-distribution generalization, and robustness. The study includes evaluations in both simulated control benchmarks and real-world navigation experiments using few-to-single observation-only demonstrations.
Cite
@article{arxiv.2507.21533,
title = {Model Predictive Adversarial Imitation Learning for Planning from Observation},
author = {Tyler Han and Yanda Bao and Bhaumik Mehta and Gabriel Guo and Anubhav Vishwakarma and Emily Kang and Sanghun Jung and Rosario Scalise and Jason Zhou and Bryan Xu and Byron Boots},
journal= {arXiv preprint arXiv:2507.21533},
year = {2026}
}
Comments
Accepted at ICLR 2026