Liao et al. (Neurocomputing 128:81–87, 2014) proposed a meta-learning approach to extreme learning machine (Meta-ELM), which can obtain good generalization performance by training multiple ELMs. However, one of its open problems is overfitting when minimizing training error. In this paper, we propose an improved meta-learning model of ELM (improved Meta-ELM) to handle the problem. The improved Meta-ELM architecture is composed of some base ELMs which are error feedback incremental extreme learning machine (EFI-ELM) and the top ELM. The improved Meta-ELM includes two stages. First, each base ELM with EFI-ELM is trained on a subset of training data. Then, the top ELM learns with the base ELMs as hidden nodes. Simulation results on some artificial and benchmark datasets show that the proposed improved Meta-ELM model is more feasible and effective than Meta-ELM.
Meta-ELM Overfitting EFI-ELM Heterogeneous
This is a preview of subscription content, log in to check access
Compliance with ethical standards
Conflict of interest
The authors (Weidong Zou, Fenxi Yao, Baihai Zhang, Zixiao Guan) of paper (Title: Improved Meta-ELM with error feedback incremental ELM as hidden nodes, NCAA-D-16-00405-R2) declare that there is no conflict of interests.
Liu HP, Liu YH, Sun FC (2015) Robust exemplar extraction using structured sparse coding. IEEE Trans Neural Netw Learn Syst 26(8):1816–1821MathSciNetCrossRefGoogle Scholar
Liu HP, Qin J, Sun FC, Di Guo Extreme kernel sparse learning for tactile object recognition, IEEE Trans Cybern, In pressGoogle Scholar
Yang YM, Jonathan Wu QM (2009) Mutilayer extreme learning machine with subnetwork nodes for representation learning. IEEE Trans Cybern 20(8):1352–1357Google Scholar
Cao J, Zhao Y, Lai X, Ong MEH, Yin C, Koh Z, Liu N (2015) Landmark recognition with sparse representation classification and extreme learning machine. J Frankl Inst 352(10):4528–4545MathSciNetCrossRefGoogle Scholar
Huang GB, Bai Z, Kasun LLC, Vong CM (2015) Local receptive fields based extreme learning machine. IEEE Comput Intell Mag 10(2):18–29CrossRefGoogle Scholar
Guo D, Zhang Y, Xiao Z, Mao M, Liu J (2015) Common nature of learning between Bp-Type and Hopfield-type neural networks. Neurocomputing 167:578–586CrossRefGoogle Scholar
Qi XX, Yuan ZH, Han XW (2015) Diagnosis of misalignment faults by tacholess order tracking analysis and RBF networks. Neurocomputing 169:439–448CrossRefGoogle Scholar
Ekici S, Yildirim S, Poyraz M (2009) A transmission line fault locator based on Elman recurrent networks. Appl Soft Comput 9(1):341–347CrossRefGoogle Scholar
Huang GB, Chen L, Siew CK (2006) Universal approximation using incremental constructive feedforward networks with random hidden nodes. IEEE Trans Neural Netw 17(4):879–892CrossRefGoogle Scholar