Document Detail


Neural-Fitted TD-Leaf Learning for Playing Othello With Structured Neural Networks.
MedLine Citation:
PMID:  24808066     Owner:  NLM     Status:  In-Data-Review    
Abstract/OtherAbstract:
This paper describes a methodology for quickly learning to play games at a strong level. The methodology consists of a novel combination of three techniques, and a variety of experiments on the game of Othello demonstrates their usefulness. First, structures or topologies in neural network connectivity patterns are used to decrease the number of learning parameters and to deal more effectively with the structural credit assignment problem, which is to change individual network weights based on the obtained feedback. Furthermore, the structured neural networks are trained with the novel neural-fitted temporal difference (TD) learning algorithm to create a system that can exploit most of the training experiences and enhance learning speed and performance. Finally, we use the neural-fitted TD-leaf algorithm to learn more effectively when look-ahead search is performed by the game-playing program. Our extensive experimental study clearly indicates that the proposed method outperforms linear networks and fully connected neural networks or evaluation functions evolved with evolutionary algorithms.
Authors:
S van den Dries; M A Wiering
Related Documents :
24128406 - Considerations for community-based mhealth initiatives: insights from three beacon comm...
23770696 - Explicit feedback maintains implicit knowledge.
25135266 - Effect of repetitive feedback on residents' communication skills improvement.
1960646 - Image, issues, and ideology: the processing of information about political candidates.
1890986 - Current tobacco, alcohol, marijuana, and cocaine use among high school students--united...
18831506 - Cost comparison of robotic, video-assisted thoracic surgery and thoracotomy approaches ...
Publication Detail:
Type:  Journal Article    
Journal Detail:
Title:  IEEE transactions on neural networks and learning systems     Volume:  23     ISSN:  2162-2388     ISO Abbreviation:  IEEE Trans Neural Netw Learn Syst     Publication Date:  2012 Nov 
Date Detail:
Created Date:  2014-05-08     Completed Date:  -     Revised Date:  -    
Medline Journal Info:
Nlm Unique ID:  101616214     Medline TA:  IEEE Trans Neural Netw Learn Syst     Country:  United States    
Other Details:
Languages:  eng     Pagination:  1701-13     Citation Subset:  IM    
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine


Previous Document:  Robust support vector regression for uncertain input and output data.
Next Document:  Global tracking control of strict-feedback systems using neural networks.