1 AIT Asian Institute of Technology

Concept generalization with neural network

AuthorWar War Khaing
Call NumberAIT Thesis no. CS-92-33
Subject(s)Neural networks (Computer science)

NoteA thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering, School of Engineering and Technology
PublisherAsian Institute of Technology
AbstractThis work concerns layered, feed-forward network, learning a classification task by backpropagation. Backpropagation was perfumed on-line in each training cycle and all processing was analog. Larger networks are trained and redundant units are removed to produce smaller network which can perform the same classification task as larger network. Two methods of pruning : Stage I and Stage II are used. Two test for generalization are used -pattern which does not have in the training set and noise distorted pattern in training set. Performance of generalization are measured as total sum square error for all pattern.
Year1992
TypeThesis
SchoolSchool of Engineering and Technology (SET)
DepartmentDepartment of Information and Communications Technologies (DICT)
Academic Program/FoSComputer Science (CS)
Chairperson(s)Sadananda, Ramakoti
Examination Committee(s)Nagarur, Nagendra N. ;Surapong Auwatanamongkol
Scholarship Donor(s)Royal Government of Thailand ;
DegreeThesis (M.Eng.) - Asian Institute of Technology, 1992


Usage Metrics
View Detail0
Read PDF0
Download PDF0