Collaborative Linear Bandits with Adversarial Agents: Near-Optimal Regret Bounds

We consider a linear stochastic bandit problem involving \(M\) agents that can collaborate via a central server to minimize regret. A fraction \(\alpha\) of these agents are adversarial and can act arbitrarily, leading to the following tension: while collaboration can potentially reduce regret, it c...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	arXiv.org 2022-06
Hauptverfasser:	Mitra, Aritra, Adibi, Arman, Pappas, George J, Hassani, Hamed
Format:	Artikel
Sprache:	eng
Schlagworte:	Algorithms Collaboration Confidence intervals Information theory Lower bounds Robustness
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!