ICML 2021spotlight11 citations
Pure Exploration and Regret Minimization in Matching Bandits
Flore Sentenac, Jialin Yi, Clement Calauzenes, Vianney Perchet, Milan Vojnovic
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).
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).}
}