Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
408135 | Neurocomputing | 2014 | 6 Pages |
Abstract
A learning scheme based on Extreme Learning Machine (ELM) and L1/2L1/2 regularization is proposed for a double parallel feedforward neural network. ELM has been widely used as a fast learning method for feedforward networks with a single hidden layer. A key problem for ELM is the choice of the (minimum) number of the hidden nodes. To resolve this problem, we propose to combine the L1/2L1/2 regularization method, that becomes popular in recent years in informatics, with ELM. It is shown in our experiments that the involvement of the L1/2L1/2 regularizer in DPFNN with ELM results in less hidden nodes but equally good performance.
Keywords
Related Topics
Physical Sciences and Engineering
Computer Science
Artificial Intelligence
Authors
Atlas Khan, Jie Yang, Wei Wu,