arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2306.04498cs.LGcs.CYcs.DC
收藏

近优隐私保护学习用于最大-最小公平多智能体老虎机

Near-Optimal Privacy-Preserving Learning for Max-Min Fair Multi-Agent Bandits

Faculty of Engineering, Bar-Ilan University(巴伊兰大学工程学院)

Amir Leshem

AI总结:

本文提出一种分布式算法,在未知支持下实现最大-最小公平分配,保持每个智能体的奖励样本和经验估计本地化,达到O(N³f(logT))的 regret,同时保留近对数依赖于T 的性能。

摘要:

我们研究在仅碰撞协调下公平的多智能体多臂老虎机学习问题。智能体在学习过程中不能显式通信,只能观察自身的奖励和多个智能体访问同一臂时碰撞的发生情况。目标是学习一个最大-最小公平分配,同时保持每个智能体的奖励样本和经验估计本地化。我们提出了一种完全分布式算法,适用于有界奖励和未知支持的情况,达到 regret O(N³f(logT)),其中 f 是任何非递减发散函数,满足 f(k-1)/f(k)→1。该算法结合了分布式智能体排序、累积轮询探索、端点重新验证的暖启动二分法以及基于碰撞的分布式拍卖用于阈值可行性测试。与基于领导者的最优算法不同,没有智能体收集其他智能体的奖励观察、经验估计或偏好。因此,该协议在避免奖励共享的操作意义上保护了奖励隐私,同时仅通过碰撞结果进行协调。与之前针对最大-最小公平老虎机的隐私保护算法相比,后者在智能体数量上具有指数依赖性,而我们的方法在保持近对数依赖于 T 的性能的同时,实现了多项式 N³ 的依赖性。分析利用了累积经验估计的集中性和端点重新验证二分法的稳定性。仿真验证了预测的与时间 horizon、智能体数量和最大-最小差距相关的缩放性能。

英文摘要:

We study fair multi-agent multi-armed bandit learning under collision-only coordination. Agents cannot communicate explicitly during learning and observe only their own rewards and whether collisions occur when several agents access the same arm. The goal is to learn a max-min fair allocation while keeping each agent's reward samples and empirical reward estimates local. We propose a fully distributed algorithm for bounded rewards with unknown support, achieving regret $O\!\left(N^3 f(\log T)\log T\right)$, where $f$ is any nondecreasing diverging function satisfying $f(k-1)/f(k)\to 1$. The algorithm combines distributed agent ordering, cumulative round-robin exploration, endpoint-revalidated warm-started bisection, and a collision-based distributed auction for threshold-feasibility tests. Unlike leader-based optimal algorithms, no agent collects the reward observations, empirical estimates, or preferences of the others. Thus, the protocol preserves reward privacy in the operational sense of avoiding reward sharing, while coordinating only through collision outcomes. Compared with previous privacy-preserving algorithms for max--min fair bandits, which have exponential dependence on the number of agents, our method achieves polynomial $N^3$ dependence while retaining near-logarithmic dependence on $T$. The analysis uses concentration of cumulative empirical estimates and stability of endpoint-revalidated bisection. Simulations confirm the predicted scaling with horizon, number of agents, and max--min gap across representative numerical settings.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑