IMPACT OF WEIGHT INITIALIZATION ON TRAINING OF SIGMOIDAL FFANN
Journal Title: ICTACT Journal on Soft Computing - Year 2018, Vol 8, Issue 3
Abstract
During training one of the most important factor is weight initialization that affects the training speed of the neural network. In this paper we have used random and Nguyen-Widrow weight initialization along with the proposed weight initialization methods for training the FFANN. We have used various types of data sets as input. Five data sets are taken from UCI machine learning repository. We have used PROP Back-Propagation algorithms for training and testing. We have taken different number of inputs and hidden layer nodes with single output node for experimentation. We have found that in almost all the cases the proposed weight initialization method gives better results.
Authors and Affiliations
Bhatia M P S, Pravin Chandra
Keywords
Related Articles
- EP ID EP532533
- DOI 10.21917/ijsc.2018.0236
- Views 68
- Downloads 0
LOSSLESS TEXT COMPRESSION FOR UNICODE TAMIL DOCUMENTS
Data compressions for different world languages including Indian languages are in high need and demand. Tamil language is one of the longest-surviving classical languages in the world. Usage of Tamil language for communi...
AN ENSEMBLE APPROACH FOR SENTIMENT CLASSIFICATION: VOTING FOR CLASSES AND AGAINST THEM
Sentiment denotes a person's opinion or feeling towards a subject that they are discussing about in that conversation. This has been one of the most researched and industrially promising fields in natural language proces...
INDEXING AND QUERY PROCESSING TECHNIQUES IN SPATIO-TEMPORAL DATA
Indexing and query processing is an emerging research field in spatio - temporal data. Most of the real-time applications such as location based services, fleet management, traffic prediction and radio frequency identifi...
ANALYSE THE PERFORMANCE OF ENSEMBLE CLASSIFIERS USING SAMPLING TECHNIQUES
In Ensemble classifiers, the Combination of multiple prediction models of classifiers is important for making progress in a variety of difficult prediction problems. Ensemble of classifiers proved potential in getting hi...
FEATURE BASED COMMUNITY DETECTION BY EXTRACTING FACEBOOK PROFILE DETAILS
FEATURE BASED COMMUNITY DETECTION BY EXTRACTING FACEBOOK PROFILE DETAILS