Entropy Learning in Neural Network

Geok See Ng (Intelligent System Laboratory, School of Computer Engineering, Nanyang Technological University, Blk N4 #2A-32, Nanyang Avenue, Singapore 639798)
D. Shi (Intelligent System Laboratory, School of Computer Engineering, Nanyang Technological University, Blk N4 #2A-32, Nanyang Avenue, Singapore 639798)
A. Wahab (Intelligent System Laboratory, School of Computer Engineering, Nanyang Technological University, Blk N4 #2A-32, Nanyang Avenue, Singapore 639798)
H. Singh (Intelligent System Laboratory, School of Computer Engineering, Nanyang Technological University, Blk N4 #2A-32, Nanyang Avenue, Singapore 639798)

ASEAN Journal on Science and Technology for Development
Vol. 20 No. 3-4 (2003): ASEAN Journal on Science and Technology for Development (AJSTD)

In this paper, entropy term is used in the learning phase of a neural network.  As learning progresses, more hidden nodes get into saturation.  The early creation of such hidden nodes may impair generalisation.  Hence entropy approach is proposed to dampen the early creation of such nodes.  The entropy learning also helps to increase the importance of relevant nodes while dampening the less important nodes.  At the end of learning, the less important nodes can then be eliminated to reduce the memory requirements of the neural network.