TY - GEN
T1 - Boosting for Online Convex Optimization
AU - Hazan, Elad
AU - Singh, Karan
N1 - Publisher Copyright:
Copyright © 2021 by the author(s)
PY - 2021
Y1 - 2021
N2 - We consider the decision-making framework of online convex optimization with a very large number of experts. This setting is ubiquitous in contextual and reinforcement learning problems, where the size of the policy class renders enumeration and search within the policy class infeasible. Instead, we consider generalizing the methodology of online boosting. We define a weak learning algorithm as a mechanism that guarantees multiplicatively approximate regret against a base class of experts. In this access model, we give an efficient boosting algorithm that guarantees near-optimal regret against the convex hull of the base class. We consider both full and partial (a.k.a. bandit) information feedback models. We also give an analogous efficient boosting algorithm for the i.i.d. statistical setting. Our results simultaneously generalize online boosting and gradient boosting guarantees to contextual learning model, online convex optimization and bandit linear optimization settings.
AB - We consider the decision-making framework of online convex optimization with a very large number of experts. This setting is ubiquitous in contextual and reinforcement learning problems, where the size of the policy class renders enumeration and search within the policy class infeasible. Instead, we consider generalizing the methodology of online boosting. We define a weak learning algorithm as a mechanism that guarantees multiplicatively approximate regret against a base class of experts. In this access model, we give an efficient boosting algorithm that guarantees near-optimal regret against the convex hull of the base class. We consider both full and partial (a.k.a. bandit) information feedback models. We also give an analogous efficient boosting algorithm for the i.i.d. statistical setting. Our results simultaneously generalize online boosting and gradient boosting guarantees to contextual learning model, online convex optimization and bandit linear optimization settings.
UR - http://www.scopus.com/inward/record.url?scp=85137709098&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85137709098&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85137709098
T3 - Proceedings of Machine Learning Research
SP - 4140
EP - 4149
BT - Proceedings of the 38th International Conference on Machine Learning, ICML 2021
PB - ML Research Press
T2 - 38th International Conference on Machine Learning, ICML 2021
Y2 - 18 July 2021 through 24 July 2021
ER -