Comments on "Local coupled extreme learning machine". Issue 3 (March 2017)

• Record Type: Journal Article
• Authors: Yu, Wan-guo; Zhou, Xu; He, Yu-lin; Ashfaq, Rana
• Is Part Of: Neural computing & applications. Volume 28:Issue 3(2017)
• Page Start: 631
• Page End: 634
• Publication Date: 2017-03
• Subjects: ELM -- LCELM -- LCFNN -- Window function; Neural networks (Computer science) -- Periodicals; Neural circuitry -- Periodicals; Artificial intelligence -- Periodicals; Neural Networks (Computer) -- Periodicals; Réseaux neuronaux (Informatique) -- Périodiques; Réseaux nerveux -- Périodiques; Intelligence artificielle -- Périodiques; 006.32
• DOI: 10.1007/s00521-016-2497-z
• Languages: English
• ISSNs: 0941-0643

Abstract

Abstract Local coupled extreme learning machine (LCELM) (Qu in Neural Comput Appl 27(1):27–33, 2016) is a recently developed variant of ELM and uses the paradigm of local coupled feed-forward neural network (LCFNN) (Sun in Neural Netw 23(1):108–113, 2010) to train an ELM model. This comment is to point out that the window function used in LCELM does not satisfy the necessary conditions of window function required by LCFNN. As a result, LCELM cannot effectively control the activation of hidden-layer node for different instances. It is possible that the improper window function can cause LCELM has the same discriminant with the basic ELM.