2024-06-11-kolev24a.md

title	abstract	layout	series	publisher	issn	id	month	tex_title	firstpage	lastpage	page	order	cycles	bibtex_author	author	date	address	container-title	volume	genre	issued	pdf	extras
Efficient imitation learning with conservative world models	We tackle the problem of policy learning from expert demonstrations without a reward function. A central challenge in this space is that these policies fail upon deployment due to issues of distributional shift, environment stochasticity, or compounding errors. Adversarial imitation learning alleviates this issue but requires additional on-policy training samples for stability, which presents a challenge in realistic domains due to inefficient learning and high sample complexity. One approach to this issue is to learn a world model of the environment, and use synthetic data for policy training. While successful in prior works, we argue that this is sub-optimal due to additional distribution shifts between the learned model and the real environment. Instead, we re-frame imitation learning as a fine-tuning problem, rather than a pure reinforcement learning one. Drawing theoretical connections to offline RL and fine-tuning algorithms, we argue that standard online world model algorithms are not well suited to the imitation learning problem. We derive a principled conservative optimization bound and demonstrate empirically that it leads to improved performance on two very challenging manipulation environments form high-dimensional raw pixel observations. We set a new state-of-the-art performance on the Franka Kitchen environment from images, requiring only 10 demos on no reward labels, as well as solving a complex dexterity manipulation task.	inproceedings	Proceedings of Machine Learning Research	PMLR	2640-3498	kolev24a	0	Efficient imitation learning with conservative world models	1777	1790	1777-1790	1777	false	Kolev, Victor and Rafailov, Rafael and Hatch, Kyle and Wu, Jiajun and Finn, Chelsea	given family Victor Kolev given family Rafael Rafailov given family Kyle Hatch given family Jiajun Wu given family Chelsea Finn	2024-06-11		Proceedings of the 6th Annual Learning for Dynamics & Control Conference	242	inproceedings	date-parts 2024 6 11	https://proceedings.mlr.press/v242/kolev24a/kolev24a.pdf