Marius Oancea
Universitatii Lucian Blaga, Sibiu, Romania
Ciprian Candea
Universitatii Lucian Blaga, Sibiu, Romania
Daniel Volovici
Universitatii Lucian Blaga, Sibiu, Romania
Ladda ner artikel
Ingår i: RobocCup-99 Team Descriptions. Simulation League
Linköping Electronic Conference Proceedings 4:31, s. 139-143
Linköping Electronic Articles in Computer and Information Science vol. 4 4:31, p. 139-143
Publicerad: 1999-12-15
ISBN:
ISSN: 1650-3686 (tryckt), 1650-3740 (online)
We use simulated soccer to study multi-agent learning. Each team member tries to learn from the corresponding human player in a real game. Following a unified approach; strategic and tactical behavior is learned synergistically by training a feed-forward neural network (ANN) with a modified back-propagation algorithm. It aims at decreasing the learning time and avoiding the local maximums. We tried to minimize the computation effort; as required in classic back-propagation (BKP) methods.
[1] Fischer; K. (1998). An Agent-Based Approach to Holonic Manufacturing Systems. In Camarinha-Matos; L.M.; Afsarmanesh; A.; and Marik;V. (Eds.); Intelligent Systems for Manufacturing. Multi-Agent Systems and Virtual Organizations. Kluwer Academic Publishers; Boston. p. 3-12
[2] Hiroaki Kitano; Minoru Asada; Yaso Kuniyoshi; Itsuki Noda; Eiichi Osaw. Robocup : The Robot World Cup Initiative
[3] Peter Stone. Layered Learning in Multi-Agent System. PhD Thesis; December 1998
[4] DARPA Neural Network Study (October; 1987-February; 1989). MIT Lincoln Lab
[5] Learning internal representations by error propagation by Rumelhart; Hinton and Williams (1986).
