期刊
IEEE TRANSACTIONS ON CONTROL OF NETWORK SYSTEMS
卷 4, 期 2, 页码 246-258出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TCNS.2015.2497098
关键词
Distributed control systems; game theory; learning algorithms; log-linear learning; multiagent systems
资金
- AFOSR [FA9550-12-1-0359]
- ONR [N00014-12-1-0643]
- National Science Foundation [ECCS-1351866]
- NASA Aeronautics scholarship program
- Philanthropic Educational Organization
- Zonta International Amelia Earhart fellowship program
- Directorate For Engineering [1638214] Funding Source: National Science Foundation
- Div Of Electrical, Commun & Cyber Sys [1638214] Funding Source: National Science Foundation
Log-linear learning has been extensively studied in the game-theoretic and distributed control literature. It is appealing for many applications because it often guarantees that the agents' collective behavior will converge in probability to the optimal system configuration. However, the worst case convergence time can be prohibitively long, that is, exponential in the number of players. Building upon the 2010 work of Shah and Shin, we formalize a modified log-linear learning algorithm whose worst case convergence time is roughly linear in the number of players. We prove this characterization for a class of potential games where agents' utility functions can be expressed as a function of aggregate behavior within a finite collection of populations. Finally, we show that the convergence time remains roughly linear in the number of players even when the players are permitted to enter and exit the game over time.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据