Announcement for Downloading full text filePlease respect the Copyright Act.
All digital full text dissertation and theses from this website are authorized the copyright owners. These copyrighted full-text dissertation and theses can be only used for academic, research and non-commercial purposes. Users of this website can search, read, and print for personal usage. In respect of the Copyright Act of the Republic of China, please do not reproduce, distribute, change, or edit the content of these dissertations and theses without any permission. Please do not create any work based upon a pre-existing work by reproduction, Adaptation, Distribution or other means.
URN etd-0712104-194000 Statistics This thesis had been viewed 3138 times. Download 992 times. Author Kuan-Zung Chen Author's Email Address No Public. Department Electrical Engineering Year 2003 Semester 2 Degree Master Type of Document Master's Thesis Language English Page Count 42 Title Using credit assignment and GBF with dynamic learning rate to enhance the ability of low-dimensional CMAC Keyword CMAC CMAC Abstract Conventional CMAC has a memory size problem at high dimensional input space. Using the neural network structure composed of small CMACs can efficiently solve this problem. By using the advantage of credit assignment and Gaussian basis function, we use it to improve the performance of the neural network structure composed of small CMACs. The simulation result shows that it still has weakness.
The dynamic learning rate and repeat training are the concepts from conventional CMAC. We use the dynamic learning rate and repeat training to overcome the weakness. The simulation shows that dynamic learning rate indeed have better performance, and dynamic learning rate with Gaussian basis function could achieve the best result.
Advisor Committee Ta-Hsiung Hung - advisor
Hung-Ching Lu - co-chair
Ming-Feng Yeh - co-chair
Files Date of Defense 2004-06-28 Date of Submission 2004-07-12