An empirical study on improving the speed and generalization of neural networks using a parallel circuit approach

dc.contributor.author	Phan, Kien Tuong
dc.contributor.author	Maul, Tomas Henrique
dc.contributor.author	Vu, Tuong Thuy
dc.date.issued	2016
dc.description.abstract	One of the common problems of neural networks, especially those with many layers, consists of their lengthy training time. We attempt to solve this problem at the algorithmic level, proposing a simple parallel design which is inspired by the parallel circuits found in the human retina. To avoid large matrix calculations, we split the original network vertically into parallel circuits and let the backpropagation algorithm flow in each subnetwork independently. Experimental results have shown the speed advantage of the proposed approach but also point out that this advantage is affected by multiple dependencies. The results also suggest that parallel circuits improve the generalization ability of neural networks presumably due to automatic problem decomposition. By studying network sparsity, we partly justified this theory and proposed a potential method for improving the design.
dc.format	Pp. 780–796
dc.identifier.uri	https://thuvienso.hoasen.edu.vn/handle/123456789/10895
dc.language.iso	en
dc.source	International Journal of Parallel Programming. Volume 45
dc.subject	Neural networks
dc.subject	Parallel circuits
dc.subject	Problem decomposition
dc.subject	Backpropagation
dc.title	An empirical study on improving the speed and generalization of neural networks using a parallel circuit approach
dc.type	Article

Files

Now showing 1 - 1 of 1

Now showing 1 - 1 of 1