Saved in:
Bibliographic Details
Main Authors: Ledford, Michael, Regli, William
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.15856
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918064035987456
author Ledford, Michael
Regli, William
author_facet Ledford, Michael
Regli, William
contents Cooperative multi-agent systems often face tasks that require coordinated actions under uncertainty. While multi-armed bandit (MAB) problems provide a powerful framework for decentralized learning, most prior work assumes individually attainable rewards. We address the challenging setting where rewards are threshold-activated: an arm yields a payoff only when a minimum number of agents pull it simultaneously, with this threshold unknown in advance. Complicating matters further, some arms are decoys - requiring coordination to activate but yielding no reward - introducing a new challenge of wasted joint exploration. We introduce Threshold-Coop-UCB (T-Coop-UCB), a decentralized algorithm that enables agents to jointly learn activation thresholds and reward distributions, forming effective coalitions without centralized control. Empirical results show that T-Coop-UCB consistently outperforms baseline methods in cumulative reward, regret, and coordination metrics, achieving near-Oracle performance. Our findings underscore the importance of joint threshold learning and decoy avoidance for scalable, decentralized cooperation in complex multi-agent
format Preprint
id arxiv_https___arxiv_org_abs_2506_15856
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Learning to Coordinate Under Threshold Rewards: A Cooperative Multi-Agent Bandit Framework
Ledford, Michael
Regli, William
Multiagent Systems
Cooperative multi-agent systems often face tasks that require coordinated actions under uncertainty. While multi-armed bandit (MAB) problems provide a powerful framework for decentralized learning, most prior work assumes individually attainable rewards. We address the challenging setting where rewards are threshold-activated: an arm yields a payoff only when a minimum number of agents pull it simultaneously, with this threshold unknown in advance. Complicating matters further, some arms are decoys - requiring coordination to activate but yielding no reward - introducing a new challenge of wasted joint exploration. We introduce Threshold-Coop-UCB (T-Coop-UCB), a decentralized algorithm that enables agents to jointly learn activation thresholds and reward distributions, forming effective coalitions without centralized control. Empirical results show that T-Coop-UCB consistently outperforms baseline methods in cumulative reward, regret, and coordination metrics, achieving near-Oracle performance. Our findings underscore the importance of joint threshold learning and decoy avoidance for scalable, decentralized cooperation in complex multi-agent
title Learning to Coordinate Under Threshold Rewards: A Cooperative Multi-Agent Bandit Framework
topic Multiagent Systems
url https://arxiv.org/abs/2506.15856