Study on Multi-Agent Q Learning Based on Prediction

Ya Xie(1*)

(*) Corresponding author

Authors' affiliations

DOI's assignment:
the author of the article can submit here a request for assignment of a DOI number to this resource!
Cost of the service: euros 10,00 (for a DOI)


Multi-agent Q learning algorithm(model) based on prediction is proposed in order to make multi-agent has the ability of online learning, which two layers of structure are adopted, planning, prediction and Q learning etc are integrated. The system state is predicted through the adoption of prediction, which the convergence speed of learning is accelerated, at the same time planning technology is used in the model through prior knowledge to solve the partial perception problems, the adaptive learning ability of algorithm has been greatly improved by the application of prediction technology, where learning algorithm has a strong ability of on-line learning and higher learning efficiency, the successful application of this algorithm in the RoboCup shows its good learning performance.
Copyright © 2013 Praise Worthy Prize - All rights reserved.


Q Learning; Online Learning; Multi-Agent; Robocup

Full Text:



Littman M L, Friend-or-foe: Q-learning in general-sum games, In 18th International Conference on Machine Learning. Williamstown, USA: Morgan Kaufmann Press, pp. 322–328, 2001

Yonghai Zhu, Shuyu Zhang, Pei Pei, Shugang Chen, Information Processing Model of Team Learning and Memory and Technical Support, (2012) International Review on Computers and Software (IRECOS), 7 (4), pp. 1812-1818.

W. Zemzem, M. Tagina, A New Approach for Reinforcement Learning in non Stationary Environment Navigation Tasks, (2012) International Review on Computers and Software (IRECOS), 7 (5), pp. 2078-2087.

Ong,S. C. W., Png, S. W., Hsu, D., and Lee, W. S, Planning under uncertainty for robotic tasks with mixed observability, IJRR, Vol. 29 N. 8, pp.1053-1068, 2010.

LI Yi, JI Hong, Q-learning for dynamic channel assignment in cognitive wireless local area network with fibre-connected distributed antennas, The Journal of China Universities of Posts and Telecommunications, Vol. 19 N. 4, pp.80-85, 2012.

Xiangfen Ji, Zhu Qi, Zhao Su, Spectrum allocation based on Q-Learning algorithm in Femtocell networks, Proceedings of 2012 IEEE International Conference on Computer Science and Automation Engineering, China, pp.782-787, 2012.

Hu yugang, The Research of Q Learning-Based Estimation of Distribution Algorithm, Proceedings of the 2011 10th International Symposium on Distributed Computing and Application to Business, Engineering and Science, China, pp.235-241,2011.

Xueqing SUN,TaoMao, Laura RAY, Dongqing SHI, Jerald KRALik, Hierarchical state-abstracted and socially augmented Q-Learning for reducing complexity in agent-based learning, Journal of Control Theory and Applications, Vol. 9 N. 3, pp. 440-445, 2011.

Sajid M. Sheikh, Shedden Masupe, M-Learning: User Expectations and Acceptance Study at the University of Botswana, (2012) International Review on Computers and Software (IRECOS), 7 (5), pp. 2183-2189.

Stone P. Layered learning in multiagent: A winning approach to robotics soccer. MIT Press, pp.125-130 ,2000.

Weiss G. Planning and learning together. In 4th International Conference on Autonomous Agent. Barcelona, Spain: ACM Press, pp.102-103,2000.

Takahashi Y, Edazawa K, and Asada M. Multi-Module learning system for behavior acquisition in multi-agent environment. In IEEE/RSJ International Conference on Intelligent Robots and Systems, pp. 927-931, 2002.


  • There are currently no refbacks.

Please send any question about this web site to
Copyright © 2005-2024 Praise Worthy Prize