Optimal Contextual Bandits with Knapsacks under Realizability via Regression Oracles

We study the stochastic contextual bandit with knapsacks (CBwK) problem, where each action, taken upon a context, not only leads to a random reward but also costs a random resource consumption in a vector form. The challenge is to maximize the total reward without violating the budget for each resou...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	arXiv.org 2023-02
Hauptverfasser:	Han, Yuxuan, Zeng, Jialin, Wang, Yang, Yang, Xiang, Zhang, Jiheng
Format:	Artikel
Sprache:	eng
Schlagworte:	Algorithms Linear functions Optimization Regression
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!