DocumentCode :
64154
Title :
Distributed Multi-Agent Online Learning Based on Global Feedback
Author :
Jie Xu ; Tekin, Cem ; Zhang, Simpson ; Van der Schaar, Mihaela
Author_Institution :
Dept. of Electr. Eng., Univ. of California Los Angeles, Los Angeles, CA, USA
Volume :
63
Issue :
9
fYear :
2015
fDate :
1-May-15
Firstpage :
2225
Lastpage :
2238
Abstract :
In this paper, we develop online learning algorithms that enable the agents to cooperatively learn how to maximize the overall reward in scenarios where only noisy global feedback is available without exchanging any information among themselves. We prove that our algorithms´ learning regrets-the losses incurred by the algorithms due to uncertainty-are logarithmically increasing in time and thus the time average reward converges to the optimal average reward. Moreover, we also illustrate how the regret depends on the size of the action space, and we show that this relationship is influenced by the informativeness of the reward structure with regard to each agent´s individual action. When the overall reward is fully informative, regret is shown to be linear in the total number of actions of all the agents. When the reward function is not informative, regret is linear in the number of joint actions. Our analytic and numerical results show that the proposed learning algorithms significantly outperform existing online learning solutions in terms of regret and learning speed. We illustrate how our theoretical framework can be used in practice by applying it to online Big Data mining using distributed classifiers.
Keywords :
Big Data; data mining; distributed algorithms; learning (artificial intelligence); multi-agent systems; pattern classification; distributed classifiers; distributed multiagent online learning algorithm; global feedback; learning speed; online Big Data mining; optimal average reward; reward informativeness; reward structure; Algorithm design and analysis; Big data; Data mining; Joints; Multi-agent systems; Noise; Signal processing algorithms; Big Data mining; distributed cooperative learning; multiagent learning; multiarmed bandits; online learning; reward informativeness;
fLanguage :
English
Journal_Title :
Signal Processing, IEEE Transactions on
Publisher :
ieee
ISSN :
1053-587X
Type :
jour
DOI :
10.1109/TSP.2015.2403288
Filename :
7041172
Link To Document :
بازگشت