English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We study the problem of stochastic bandits with adversarial corruptions inthe cooperative multi-agent setting, where $V$ agents interact with a common$K$-armed bandit problem, and each pair of agents can communicate with eachother to expedite the learning process. In the problem, the rewards areindependently sampled from distributions across all agents and rounds, but theymay be corrupted by an adversary. Our goal is to minimize both the overallregret and communication cost across all agents. We first show that an additiveterm of corruption is unavoidable for any algorithm in this problem. Then, wepropose a new algorithm that is agnostic to the level of corruption. Ouralgorithm not only achieves near-optimal regret in the stochastic setting, butalso obtains a regret with an additive term of corruption in the corruptedsetting, while maintaining efficient communication. The algorithm is alsoapplicable for the single-agent corruption problem, and achieves a highprobability regret that removes the multiplicative dependence of $K$ oncorruption level. Our result of the single-agent case resolves an open questionfrom Gupta et al. [2019].

Cooperative Stochastic Multi-agent Multi-armed Bandits Robust to Adversarial Corruptions