Modified Q-Learning for Intelligent System 


Vol. 33,  No. 2, pp. 82-87, Feb.  2008


PDF Full-Text
  Abstract

In this paper, for a continuous state space applications, a novel method of Q-learning is proposed, where the method incorporates a region-based reward assignment being used to solve structural credit assignment problem and a convex clustering approach to find a region with the same reward attribution property. Our learning method can estimate a current Q-value of an arbitrarily given state by using effect functions, and has the ability to learn its action ssimilar to that of Q-learning. Thus, our method enables a system to adapt smoothly to a real environment. To show the validity of our method, the proposed Q-learning method is compared with conventional Q-learning method through a simple two dimensional free space navigation problem.

  Statistics
Cumulative Counts from November, 2022
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]

Y. J. Kim, "Modified Q-Learning for Intelligent System," The Journal of Korean Institute of Communications and Information Sciences, vol. 33, no. 2, pp. 82-87, 2008. DOI: .

[ACM Style]

Young Jun Kim. 2008. Modified Q-Learning for Intelligent System. The Journal of Korean Institute of Communications and Information Sciences, 33, 2, (2008), 82-87. DOI: .

[KICS Style]

Young Jun Kim, "Modified Q-Learning for Intelligent System," The Journal of Korean Institute of Communications and Information Sciences, vol. 33, no. 2, pp. 82-87, 2. 2008.

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880