Dissertation > Excellent graduate degree dissertation topics show
The Realization of Weight Function under MATLAB
Author: PanRong
Tutor: ZhangDaiYuan
School: Nanjing University of Posts and Telecommunications
Course: Computer technology
Keywords: B-spline weight function Padé weight function The rational spline rights function Orthogonal weight function Rational weight function MATLAB implementation
CLC: TP319
Type: Master's thesis
Year: 2012
Downloads: 47
Quote: 0
Read: Download Dissertation
Abstract
Traditional neural network learning algorithms ( such as BP algorithm , RBF algorithm) , the weight is constant and can not be completely memory sample information . Response to this problem , Nanjing Posts and Telecommunications University 's Professor Zhang on behalf of Yuan in its exclusively with \and in this new network algorithm , neural network generalization is enhanced with the increase of the number of samples . But under normal circumstances , by a finite number of known sample point for the theoretical weight function is impossible . In the Research Professor Zhang Daiyuan based on interpolation method to obtain the approximate weight function , the the structure weight function neural network . In the theoretical part of the MATLAB programming , theoretical research is mainly based on the weight function neural network . Interpolation method to construct approximate weight function , the neural network structure of the weight function in multiple-input , single - output and multi- input , multi- output error analysis in both cases . The analysis results show that the more the number of training samples , the smaller the error value . Part of the software implementation , the main topic is about weight function neural network applications in MATLAB to achieve . Demand analysis for the software to design the right function neural network , created through the network parameters for sample collection , the weighting factor of the input design to be able to read plain text files and binary files . Inside the auxiliary functions of the software , the added weight function neural network sensitivity analysis . Meanwhile, the software has a help menu and the English version , improve the functionality of the software applications .
|
Related Dissertations
- Research of Algorithm for Objects Extraction and Identification in Machine Visual Image,TP391.41
- Clustering and Neural Network Based Intrusion Detection System,TP393.08
- The Methods for Detection and Classification of Steel Ingot Surface Defects,TP274
- County construction land intensive use of evaluation studies,TP319
- Assistant Software for Dynamic Analysis in Oilfield Development,TP319
- Study on the Specificating and Appraising System for Reliability of Civil Buildings,TP319
- The Research of the Factor Analysis and Implementation in AFC Data Analysis System,TP319
- Study of E-Learning Personalized Recommendation System,TP319
- Qualitative Analysis of Helicopter Stability under the Precipitation and Thunderstorm,TP319
- The Complete Characteristics Predictions of Pumps Based on GARBFNN,TP319
- The Research of BP Neural Network Model in Stock Price Forecasting,TP319
- The Research & Practice of Software of Pharmaceutical Producing and Supplying,TP319
- Third-party Purchasing Service System & Its Application in Medicine Purchasing Management,TP319
- A Path Searching Algorithm and It’s Optimization in Personal Navigation System,TP319
- Research and Application of Clustering Algorithm on the Power Marketing Decision Support System,TP319
- Research on Intelligent Test Paper Composition System Based on Genetic Algorithm,TP319
- Application and Research of Urban and Rural Integrative Cadastral Information System,TP319
- Shenzhen Metro AFC system for information security management construction,TP319
- XML data stream query system based on ontology,TP319
- Research on the Information Retrieval System of Power Grid and Typhoon Based on WebGIS,TP319
CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer software > Specific applications
© 2012 www.DissertationTopic.Net Mobile
|