Aug 2026· arXiv.org· Vol abs/2608.10529· 0 citations· 16 references
Computer Science
TL;DR
This work develops robust decentralized algorithms for each setting and derive regret guarantees that nearly match centralized heavy-tailed rates.
Abstract
The multi-armed bandit problem is a central framework in sequential decision-making, extensively studied under sub-Gaussian reward assumptions. However, real-world applications often involve heavy-tailed reward distributions and decentralized, information-asymmetric interactions. We study multi-agent multi-armed bandits with heavy-tailed rewards under three information-asymmetry regimes: unobserved actions with common rewards, observed actions with independent rewards, and unobserved actions with independent rewards. We develop robust decentralized algorithms for each setting and derive regret guarantees that nearly match centralized heavy-tailed rates. Experiments on a Pareto-distributed reward environment validate our theoretical findings and illustrate the trade-offs between synchronization, coordination, and exploration across the three regimes.
. It is a universal truth that classical multi-armed bandit (MAB) theory relies on the stationary assumption of reward distributions. However, this assumption is not viable in systems with multiple adaptive interacting agents where interactions endogenously redefine rewards. This paper explores bandit learning in non-s...
Bei-Shen Guo· Proceedings of the 4th Inter...· 0 citations
We study decentralized bandit learning in discounted infinite-horizon, general-sum Markov games with continuous actions. Agents observe the public state, but only their own realized rewards, and use simultaneous perturbation stochastic approximation to update their actions. Under strong monotonicity of the negative run...
This work introduces Decision-Relevant Fresh Comparison (DRFC), which uses new, balanced samples from all agents to compare arms at the network level and switches only when fresh global evidence indicates that the common best arm has changed, and proves a high-probability dynamic regret bound with no adaptation term de...
We develop an index policy for finite-horizon Bernoulli multi-armed bandits from minimax solutions to single-arm bandit (SAB) problems. Each SAB problem involves choosing between an unknown Bernoulli arm and a known reward. We show that minimizing worst-case regret of SAB problems over all non-anticipative policies adm...
Hui-Kang Liu, Zheng-Chao Wang, Daniel Kuhn et al.· arXiv.org· 0 citations
The multi-armed bandit problem models the trade-off between exploration and exploitation. In this paper, we study the variant where exploration has a hard budget and precedes exploitation. Prior guarantees include a 1/3 approximation by Guha and Munagala using Lagrangian duality, and a 4/27 approximation by Ma against...
This work analyzes a deceiver-victim pair in the minimal two-player strongly monotone setting and proves that the deceptive bandit learning dynamics converge to an arbitrarily small neighborhood of the DNE while retaining optimal convergence rates.
Michael Tang, Mahmoud Abdelgalil, J. Poveda· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.