High-dimensional neural feature design for layer-wise reduction of training cost

High-dimensional neural feature design for layer-wise reduction of training cost

Abstract We design a rectified linear unit-based multilayer neural network by mapping the feature vectors to a higher dimensional space in every layer. We design the weight matrices in every layer to ensure a reduction of the training cost as the number of layers increases. Linear projection to the...

Full description

Bibliographic Details
Main Authors: Alireza M. Javid, Arun Venkitaraman, Mikael Skoglund, Saikat Chatterjee
Format: Article
Language:English
Published: SpringerOpen 2020-09-01
Series:EURASIP Journal on Advances in Signal Processing
Subjects:
Rectified linear unit
Feature design
Neural network
Convex cost function
Online Access:http://link.springer.com/article/10.1186/s13634-020-00695-2

Internet

http://link.springer.com/article/10.1186/s13634-020-00695-2

Similar Items