A Comprehensive Evaluation of Weight Growth and Weight Elimination Methods Using the Tangent Plane Algorithm
Journal Title: International Journal of Advanced Computer Science & Applications - Year 2013, Vol 4, Issue 6
Abstract
The tangent plane algorithm is a fast sequential learning method for multilayered feedforward neural networks that accepts almost zero initial conditions for the connection weights with the expectation that only the minimum number of weights will be activated. However, the inclusion of a tendency to move away from the origin in weight space can lead to large weights that are harmful to generalization. This paper evaluates two techniques used to limit the size of the weights, weight growing and weight elimination, in the tangent plane algorithm. Comparative tests were carried out using the Extreme Learning Machine which is a fast global minimiser giving good generalization. Experimental results show that the generalization performance of the tangent plane algorithm with weight elimination is at least as good as the ELM algorithm making it a suitable alternative for problems that involve time varying data such as EEG and ECG signals.
Authors and Affiliations
P May, E Zhou, C. Lee
Keywords
Related Articles
- EP ID EP130692
- DOI 10.14569/IJACSA.2013.040621
- Views 91
- Downloads 0
Graduation Certificate Verification Model: A Preliminary Study
Graduation certificates issued by universities and other educational institutions are one of the most important documents for a graduate. It is a proof of graduate’s qualifications and can be used to advance forward in o...
Effect of Routing Protocols and Layer 2 Mediums on Bandwidth Utilization and Latency
Computer networks (CNS) are progressing as emerging field in information and communication technology (ICT). Various computer networks related problems relies on performance of computer network specifically bandwidth uti...
Comparison of Agile Method and Scrum Method with Software Quality Affecting Factors
The software industry used software development lifecycle (SDLC) to design, develop, produce high quality, reliable and cost-effective software products. To develop an application, project team used some methodology whic...
Comparative Analysis of Raw Images and Meta Feature based Urdu OCR using CNN and LSTM
Urdu language uses cursive script which results in connected characters constituting ligatures. For identifying characters within ligatures of different scales (font sizes), Convolution Neural Network (CNN) and Long Shor...
Image Encryption Technique based on the Entropy Value of a Random Block
The use of digital images in most fields of information technology systems makes these images usually contain confidential information. When these images transmitted via the Internet especially in the Cloud, it becomes n...