2024-04-18-alizadeh24a.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
Pessimistic Off-Policy Multi-Objective Optimization	Multi-objective optimization is a class of optimization problems with multiple conflicting objectives. We study offline optimization of multi-objective policies from data collected by a previously deployed policy. We propose a pessimistic estimator for policy values that can be easily plugged into existing formulas for hypervolume computation and optimized. The estimator is based on inverse propensity scores (IPS), and improves upon a naive IPS estimator in both theory and experiments. Our analysis is general, and applies beyond our IPS estimators and methods for optimizing them.	inproceedings	Proceedings of Machine Learning Research	PMLR	2640-3498	alizadeh24a	0	Pessimistic Off-Policy Multi-Objective Optimization	2980	2988	2980-2988	2980	false	Alizadeh, Shima and Bhargava, Aniruddha and Gopalswamy, Karthick and Jain, Lalit and Kveton, Branislav and Liu, Ge	given family Shima Alizadeh given family Aniruddha Bhargava given family Karthick Gopalswamy given family Lalit Jain given family Branislav Kveton given family Ge Liu	2024-04-18		Proceedings of The 27th International Conference on Artificial Intelligence and Statistics	238	inproceedings	date-parts 2024 4 18	https://proceedings.mlr.press/v238/alizadeh24a/alizadeh24a.pdf