2023-12-02-lyu23a.md

title	section	openreview	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
Task-Oriented Koopman-Based Control with Contrastive Encoder	Oral	q0VAoefCI2	We present task-oriented Koopman-based control that utilizes end-to-end reinforcement learning and contrastive encoder to simultaneously learn the Koopman latent embedding, operator, and associated linear controller within an iterative loop. By prioritizing the task cost as the main objective for controller learning, we reduce the reliance of controller design on a well-identified model, which, for the first time to the best of our knowledge, extends Koopman control from low to high-dimensional, complex nonlinear systems, including pixel-based tasks and a real robot with lidar observations. Code and videos are available: https://sites.google.com/view/kpmlilatsupp/.	inproceedings	Proceedings of Machine Learning Research	PMLR	2640-3498	lyu23a	0	Task-Oriented Koopman-Based Control with Contrastive Encoder	93	105	93-105	93	false	Lyu, Xubo and Hu, Hanyang and Siriya, Seth and Pu, Ye and Chen, Mo	given family Xubo Lyu given family Hanyang Hu given family Seth Siriya given family Ye Pu given family Mo Chen	2023-12-02		Proceedings of The 7th Conference on Robot Learning	229	inproceedings	date-parts 2023 12 2	https://proceedings.mlr.press/v229/lyu23a/lyu23a.pdf