2023-12-02-ahn23a.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
Preference learning for guiding the tree search in continuous POMDPs	Poster	09UL1dCqf2n	A robot operating in a partially observable environment must perform sensing actions to achieve a goal, such as clearing the objects in front of a shelf to better localize a target object at the back, and estimate its shape for grasping. A POMDP is a principled framework for enabling robots to perform such information-gathering actions. Unfortunately, while robot manipulation domains involve high-dimensional and continuous observation and action spaces, most POMDP solvers are limited to discrete spaces. Recently, POMCPOW has been proposed for continuous POMDPs, which handles continuity using sampling and progressive widening. However, for robot manipulation problems involving camera observations and multiple objects, POMCPOW is too slow to be practical. We take inspiration from the recent work in learning to guide task and motion planning to propose a framework that learns to guide POMCPOW from past planning experience. Our method uses preference learning that utilizes both success and failure trajectories, where the preference label is given by the results of the tree search. We demonstrate the efficacy of our framework in several continuous partially observable robotics domains, including real-world manipulation, where our framework explicitly reasons about the uncertainty in off-the-shelf segmentation and pose estimation algorithms.	inproceedings	Proceedings of Machine Learning Research	PMLR	2640-3498	ahn23a	0	Preference learning for guiding the tree search in continuous POMDPs	3929	3948	3929-3948	3929	false	Ahn, Jiyong and Son, Sanghyeon and Lee, Dongryung and Han, Jisu and Son, Dongwon and Kim, Beomjoon	given family Jiyong Ahn given family Sanghyeon Son given family Dongryung Lee given family Jisu Han given family Dongwon Son given family Beomjoon Kim	2023-12-02		Proceedings of The 7th Conference on Robot Learning	229	inproceedings	date-parts 2023 12 2	https://proceedings.mlr.press/v229/ahn23a/ahn23a.pdf