Emergence of Individuality and Sociality through Reinforcement Learning

Summary A concept of individuality and sociality is introduced as a method to avoid conflicts of individual interests in multi-agent systems. It is considered that each agent has its individuality when the conflicts are resolved by making its own mapping from the sensory input to the action output. On the other hand, each agent has sociality when the conflicts are avoided by some common input-output mapping, which is commonly called rules. A conflict avoidance task in which passengers are getting on and off a train are taken as an example, and the emergence processes of both behavioral characters are explained. Furthermore, it is shown that the differentiation of the agent into one of them is adaptively realized by reinforcement learning based on local rewards according to the asymmetry of environment, number of agents, identification of the other agents, or physical ability of agents.

Reference
3. Katsunari Shibata, Masahide Ueda and Koji Ito:
Emergence and Differentiation Model of Individuality and Sociality by Reinforcement Learning,
Trans. of SICE (The Society of Instrument and Control Engineers), Vol. 39, No. 5, pp.494--502, 2003.5 (in Japanese)
柴田克成、上田雅英、伊藤宏司:
強化学習による個性・社会性の発現・分化モデル,
計測自動制御学会論文集, Vol. 39, No. 5, pp. 494-502, 2003.5
pdf File (9 pages, 186kB)

2. Katsunari Shibata, Masahide Ueda, and Koji Ito:
Emergence of Individuality and Sociality by Reinforcement Learning,
Proc. of AROB (Int'l Symposium on Artificial Life and Robotics) 5th 2000, Vol. 2, pp. 589 - 592, 2000.1
[Individuality, Sociality, Reinforcement Learning, Multi-Agent Systems]
pdf File (4 pages, 119kB)

1. 上田雅英，柴田克成，伊藤宏司：
マルチエージェント系における個性・社会性の学習的生成,
第11回自律分散システムシンポジウム資料, pp. 299-302, 1999.1
(in Japanese)

Return to my home page (English)

Return to my home page (Japanese)