Pure Exploration and Regret Minimization in Matching Bandits

Flore Sentenac, Jialin Yi, Clement Calauzenes, Vianney Perchet, Milan Vojnovic
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:9434-9442, 2021.

Abstract

Finding an optimal matching in a weighted graph is a standard combinatorial problem. We consider its semi-bandit version where either a pair or a full matching is sampled sequentially. We prove that it is possible to leverage a rank-1 assumption on the adjacency matrix to reduce the sample complexity and the regret of off-the-shelf algorithms up to reaching a linear dependency in the number of vertices (up to to poly-log terms).

Cite this Paper

BibTeX
@InProceedings{pmlr-v139-sentenac21a, title = {Pure Exploration and Regret Minimization in Matching Bandits}, author = {Sentenac, Flore and Yi, Jialin and Calauzenes, Clement and Perchet, Vianney and Vojnovic, Milan}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, pages = {9434--9442}, year = {2021}, editor = {Meila, Marina and Zhang, Tong}, volume = {139}, series = {Proceedings of Machine Learning Research}, month = {18--24 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v139/sentenac21a/sentenac21a.pdf}, url = {https://proceedings.mlr.press/v139/sentenac21a.html}, abstract = {Finding an optimal matching in a weighted graph is a standard combinatorial problem. We consider its semi-bandit version where either a pair or a full matching is sampled sequentially. We prove that it is possible to leverage a rank-1 assumption on the adjacency matrix to reduce the sample complexity and the regret of off-the-shelf algorithms up to reaching a linear dependency in the number of vertices (up to to poly-log terms).} }
Endnote
%0 Conference Paper %T Pure Exploration and Regret Minimization in Matching Bandits %A Flore Sentenac %A Jialin Yi %A Clement Calauzenes %A Vianney Perchet %A Milan Vojnovic %B Proceedings of the 38th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Marina Meila %E Tong Zhang %F pmlr-v139-sentenac21a %I PMLR %P 9434--9442 %U https://proceedings.mlr.press/v139/sentenac21a.html %V 139 %X Finding an optimal matching in a weighted graph is a standard combinatorial problem. We consider its semi-bandit version where either a pair or a full matching is sampled sequentially. We prove that it is possible to leverage a rank-1 assumption on the adjacency matrix to reduce the sample complexity and the regret of off-the-shelf algorithms up to reaching a linear dependency in the number of vertices (up to to poly-log terms).
APA
Sentenac, F., Yi, J., Calauzenes, C., Perchet, V. & Vojnovic, M.. (2021). Pure Exploration and Regret Minimization in Matching Bandits. Proceedings of the 38th International Conference on Machine Learning, in Proceedings of Machine Learning Research 139:9434-9442 Available from https://proceedings.mlr.press/v139/sentenac21a.html.

Related Material