Online Learning with Dependent Stochastic Feedback Graphs
Corinna Cortes, Giulia Desalvo, Claudio Gentile, Mehryar Mohri, Ningshan Zhang
Abstract
A general framework for online learning with partial information is one where feedback graphs specify which losses can be observed by the learner. We study a challenging scenario where feedback graphs vary stochastically with time and, more importantly, where graphs and losses are dependent. This scenario appears in several real-world applications that we describe where the outcome of actions are correlated. We devise a new algorithm for this setting that exploits the stochastic properties of the graphs and that benefits from favorable regret guarantees. We present a detailed theoretical analysis of this algorithm, and also report the result of a series of experiments on real-world datasets, which show that our algorithm outperforms standard baselines for online learning with feedback graphs.
BibTeX
@InProceedings{pmlr-v119-cortes20b,
title = {Online Learning with Dependent Stochastic Feedback Graphs},
author = {Cortes, Corinna and Desalvo, Giulia and Gentile, Claudio and Mohri, Mehryar and Zhang, Ningshan},
booktitle = {Proceedings of the 37th International Conference on Machine Learning},
pages = {2154--2163},
year = {2020},
editor = {III, Hal Daumé and Singh, Aarti},
volume = {119},
series = {Proceedings of Machine Learning Research},
month = {13--18 Jul},
publisher = {PMLR},
pdf = {http://proceedings.mlr.press/v119/cortes20b/cortes20b.pdf},
url = {https://proceedings.mlr.press/v119/cortes20b.html},
abstract = {A general framework for online learning with partial information is one where feedback graphs specify which losses can be observed by the learner. We study a challenging scenario where feedback graphs vary stochastically with time and, more importantly, where graphs and losses are dependent. This scenario appears in several real-world applications that we describe where the outcome of actions are correlated. We devise a new algorithm for this setting that exploits the stochastic properties of the graphs and that benefits from favorable regret guarantees. We present a detailed theoretical analysis of this algorithm, and also report the result of a series of experiments on real-world datasets, which show that our algorithm outperforms standard baselines for online learning with feedback graphs.}
}