Enhanced random search based incremental extreme learning machine

By G. Huang, Lei Chen. Neurocomputing, 2008

902 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/dffb3a5cf8d7cfa687eaf2fb0070c3c9e7389bc2

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. G. Huang
2. Dong Yu
3. Li Deng
4. Shiji Song
5. Gao Huang
6. Yoshua Bengio
7. Jianfeng Gao
8. Quoc V. Le

In Electrical and Electronic Engineering

1. G. Huang
2. Dong Yu
3. Li Deng
4. Shiji Song
5. Yuan Lan
6. Dianhui Wang
7. Hongming Zhou
8. Xiaojian Ding

Citations in the corpus

• A structure optimisation algorithm for feedforward neural network construction (2013, methodology): “(3) The mean training CPU time is much shorter than the other algorithms (except the EI-ELM algorithm) for Breast Cancer, Iris, and Yeast in Case A.1.”
• Voting based extreme learning machine (2012, background): “It has been further shown [15–18] that many types of computational hidden nodes which may not be neuron alike nodes can be used in ELM as long as they are piecewise nonlinear, such as radial basis function (RBF) hidden nodes [16], fully complex nodes [18], wavelets [3,4], etc.”
• What are Extreme Learning Machines? Filling the Gap Between Frank Rosenblatt’s Dream and John von Neumann’s Puzzle (2015, methodology): “Its machine-based learning efficiency was confirmed in 2004 [24], and its universal approximation capability (for ‘‘generalized SLFNs’’ in which a hidden node may be a subnetwork of several nodes and/or with almost any nonlinear piecewise continuous neurons (although their exact mathematical modelin”
• Optimization method based extreme learning machine for classification (2010, methodology): “…www.elsevier.com/locate/neucom Optimization method based extreme learning machine for classification$ Guang-Bin Huang a, , Xiaojian Ding a,b, Hongming Zhou a a School of Electrical and Electronic Engineering, Nanyang Technological University, Nanyang Avenue, Singapore 639798, Singapore b…”
• Local Receptive Fields Based Extreme Learning Machine (2015, methodology): “Different from the common understanding of neural networks, ELM theories [9]–[11] show that although hidden neurons play critical roles, they need not be adjusted.”
• Efficient and effective algorithms for training single-hidden-layer neural networks (2012, methodology): “EI-ELM is only slightly better than ELM with an average accuracy of 94.78%.”
• Overview of model-based online state-of-charge estimation using Kalman filter family for lithium-ion batteries (2019, background): “Some machine learning tools, such as artificial neural network [90,91], fuzzy logic [92–95], support vector mechanism [83,96–99], support vector regression [23,100], extreme learning machine [101,102] and genetic algorithm (GA), are used for SOC estimation [103–105].”
• Trends in extreme learning machines: A review (2015, methodology): “…from the conventional neural networks learning tenet that hidden node parameters ( a i , b i ) need to be adjusted during training, ELM theories (Huang & Chen, 2007, 2008; Huang, Chen, et al., 2006) rigorously prove that hidden node parameters ( a i , b i ) can be randomly generated according…”
• Pruning algorithms of neural networks — a comparative study (2013, positive): “The first, also called as constructive methods, start with a minimal network and add new hidden nodes during the training process [18] [19] [20].”
• Extreme learning machines: a survey (2011, methodology): “3 can be further extended from additive or RBF hidden nodes cases to ‘‘generalized’’ SLFNs [8, 9].”
• Weighted extreme learning machine for imbalance learning (2013, background): “Extreme learning machine (ELM) [1,2,11] was originally proposed for the single-hidden layer feedforward neural networks and was then extended to the ‘‘generalized’’ single-hidden layer feedforward networks (SLFNs) where the hidden layer need not be neuron alike [3,4].”
• Extreme Learning Machine for Regression and Multiclass Classification (2012, methodology): “According to ELM learning theory, a widespread type of feature mappings h ( x ) can be used in ELM so that ELM can approximate any continuous target functions (refer to [14]–[16] for details).”