Speeding Up Back-Propagation Learning (SUBPL) Algorithm: A New Modified Back-Propagation Algorithm
Journal Title: International Journal of Engineering Research and Advanced Technology - Year 2016, Vol 2, Issue 3
Abstract
The convergence speed is the most important feature of Back-Propagation (BP) algorithm. A lot of improvements were proposed to this algorithm since its presentation, in order to speed up the convergence phase. In this paper, a new modified BP algorithm called Speeding up Back-Propagation Learning (SUBPL) algorithm is proposed and compared to the standard BP. Different data sets were implemented and experimented to verify the improvement in SUBPL.
Authors and Affiliations
Keywords
Related Articles
- EP ID EP211780
- DOI -
- Views 60
- Downloads 0
Quality Aspects of Dref-III Spun Yarns by Optimisation of Parameters with Box and Behnken’s 3x3 Model
Studies have shown the effect of two different fibre parameters (fibre length and fibre fineness) as well as core-sheath ratio of yarn on the physical properties and structure of Dref-III yarns. Different response surfac...
ADVERTISEMENT MANAGEMENT IN SOCIAL NETWORK BASED ON USER INFORMATION
An advertisement company can serve their ads depends on every user interest. They provide the ads depends on User profile and him/her activity in social Network by collecting data from user activities. The purpose of col...
DEGRADATION OF ORGANIC POLLUTANTS BY INDIRECT ELECTROCHEMICAL OXIDATION PROCESS USING COMPOSITE ELECTRODE
The aim of this research work is to investigate the efficiency of electrochemical oxidation of industrial polluted water which contains Fast Blue B salt. A novel composite electrode was fabricated by pressing together Co...
Investigation of strength characteristics of cement bagasse ash mortars using full factorial approach
The cement industry is one of the largest embodied energy consuming and CO2 emitting industry. It is found that to produce one kg of cement, it requires 4.2MJ of energy and liberates about one kg of CO2 which is one of t...
A Secure Framework for efficient search and mining over an encrypted data using homomorphic encryption
In the emerging networked environments learning tasks are encountering situations in which the relevant data exists in a number of geo-graphically distributed databases that are connected by communication networks. For t...