A Neural Scheduling Method for Efficient Use of Distributed Multi-Agent System


The Transactions of the Korea Information Processing Society (1994 ~ 2000), Vol. 7, No. 10, pp. 3105-3112, Oct. 2000
10.3745/KIPSTE.2000.7.10.3105,   PDF Download:

Abstract

The distributed agent system finds the solution of the problem given through the reciprocal cooperation between agents, but its concept and their mutual operation are complicated Tins paper suggests a neural scheduling method for an efficient job link between the agents as a solution for these problems First, a total of 8750 case patterns are composed on the base of system operation scenario for drawing up of a neural scheduling. Among these cases, we learned in 6×5×2 neural network by using 2,000 and 6750 cases for learning data and the lest pattern respectively. The result of the learning showed more than 90% of convergence rates about learning and test patterns in 200 learning times. The agent creation, its job performance, and the division and mutual cooperation between them were also made well by using the result for a neural scheduling data.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from September 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.


Cite this article
[IEEE Style]
J. P. Jeong, K. J. Lee, C. Y. Jung, "A Neural Scheduling Method for Efficient Use of Distributed Multi-Agent System," The Transactions of the Korea Information Processing Society (1994 ~ 2000), vol. 7, no. 10, pp. 3105-3112, 2000. DOI: 10.3745/KIPSTE.2000.7.10.3105.

[ACM Style]
Jong Pil Jeong, Kee Jun Lee, and Chai Yeoung Jung. 2000. A Neural Scheduling Method for Efficient Use of Distributed Multi-Agent System. The Transactions of the Korea Information Processing Society (1994 ~ 2000), 7, 10, (2000), 3105-3112. DOI: 10.3745/KIPSTE.2000.7.10.3105.