A study on residence error of training an extreme learning machine and its application to evolutionary algorithms

Fu, Aimin; Wang, Xizhao; He, Yulin; Wang, Lai-Sheng

doi:10.1016/j.neucom.2014.04.067

Cited by 17 publications

(9 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Extreme learning machine (ELM) [ 12 , 13 ] is a type of noniterative training algorithm for single hidden-layer feed-forward neural network (SLFN). ELM randomly selects the input-layer weights/hidden-layer biases and analytically calculates the output-layer weights [ 6 ]. Because ELM does not perform complex parameter adjustment (e.g., learning rate, learning epoches, stopping criteria) and time-consuming weight updates, ELM is simpler and faster than the traditional back-propagation (BP) algorithm [ 5 , 10 ].…”

Section: Introductionmentioning

confidence: 99%

A novel correlation Gaussian process regression-based extreme learning machine

Zhang

et al. 2023

Knowl Inf Syst

View full text Add to dashboard Cite

An obvious defect of extreme learning machine (ELM) is that its prediction performance is sensitive to the random initialization of input-layer weights and hidden-layer biases. To make ELM insensitive to random initialization, GPRELM adopts the simple an effective strategy of integrating Gaussian process regression into ELM. However, there is a serious overfitting problem in kernel-based GPRELM ( k GPRELM). In this paper, we investigate the theoretical reasons for the overfitting of k GPRELM and further propose a correlation-based GPRELM ( c GPRELM), which uses a correlation coefficient to measure the similarity between two different hidden-layer output vectors. c GPRELM reduces the likelihood that the covariance matrix becomes an identity matrix when the number of hidden-layer nodes is increased, effectively controlling overfitting. Furthermore, c GPRELM works well for improper initialization intervals where ELM and k GPRELM fail to provide good predictions. The experimental results on real classification and regression data sets demonstrate the feasibility and superiority of c GPRELM, as it not only achieves better generalization performance but also has a lower computational complexity.

show abstract

Section: Introductionmentioning

confidence: 99%

A novel correlation Gaussian process regression-based extreme learning machine

Zhang

et al. 2023

Knowl Inf Syst

View full text Add to dashboard Cite

show abstract

“…Extreme Learning Machine (ELM) [12,13] is a kind of non-iterative training algorithm for Single hidden-Layer Feed-forward neural Network (SLFN). ELM randomly selects the input-layer weights and hidden-layer biases and analytically calculates the output-layer weights [6]. Due to avoiding the complex parameter adjustment (e.g., learning rate, learning epoches, stopping criteria, etc) and timeconsuming weight updating, ELM is simpler and faster than the traditional Back-Propagation (BP) algorithm [5,10].…”

Section: Introductionmentioning

confidence: 99%

A Novel Correlation Gaussian Process Regression-Based Extreme Learning Machine

Yu-lin

YUAN

et al. 2022

Preprint

View full text Add to dashboard Cite

One obvious defect of Extreme Learning Machine (ELM) is that the prediction performance of ELM is sensitive to the random initialization of input-layer weights and hidden-layer biases. GPRELM integrating Gaussian Process Regression (GPR) into ELM is a newly-proposed, simple and effective strategy to make ELM insensitive to the random initialization. However, the kernel based GPRELM (kGPRELM) exists serious over-fitting. In this paper, we analyze the reason for over-fitting of kGPRELM in theory and further propose a correlation based GPRELM (cGPRELM) which uses the correlation coefficient to measure the similarity between two different hidden-layer output vectors. cGPRELM reduces the possibility that covariance matrix becomes an identity matrix with the increase of hidden-layer nodes and thus controls the over-fitting effectively. Meanwhile, cGPRELM works well for improper initialization intervals where ELM and kGPRELM can't provide the good predictions. The experimental results on real classification and regression data sets demonstrate the feasibility and superiority of cGPRELM, which not only obtains the better generalization performance but also has the lower computational complexity.

show abstract

“…After appropriate features are selected, the next step of the data-driven based RUL prediction process is to establish a prediction model. More and more artificial intelligence algorithms have been applied to predict the RUL, such as support vector machine [15][16][17], extreme learning machine [18], joint approximate diagonalization of eigenmatrices (JADE) algorithm [14], particle filter [19], deep learning [20], and so on. These models can estimate health states and update model parameters according to real-time information.…”

Section: Introductionmentioning

confidence: 99%

Remaining Useful Life Prediction of Machinery based on K-S Distance and LSTM Neural Network

Yang¹,

Guo²,

Dou³

2019

IJPE

View full text Add to dashboard Cite

The remaining useful life is key to the decision-making of machinery maintenance. The online prediction of remaining useful life has become a very urgent need for mechanical equipment with high reliability requirements. The aim of this paper is to provide a simple and effective method for predicting the remaining life of the machine under the condition of small sample. The Kolmogorov-Smirnov test theory is used to extract the health state feature of the machine. Based on the Long and Short Term Memory (LSTM) theory, an online method of remaining useful life prediction is proposed. The bearing life vibration data verification shows that the Kolmogorov-Smirnov distance is sensitive to the development and expansion of the defects. Furthermore, the proposed method of remaining useful life prediction based on LSTM theory has high prediction accuracy. The technician can then use this method to take appropriate maintenance operations.

show abstract

A study on residence error of training an extreme learning machine and its application to evolutionary algorithms

Cited by 17 publications

References 24 publications

A novel correlation Gaussian process regression-based extreme learning machine

A novel correlation Gaussian process regression-based extreme learning machine

A Novel Correlation Gaussian Process Regression-Based Extreme Learning Machine

Remaining Useful Life Prediction of Machinery based on K-S Distance and LSTM Neural Network

Contact Info

Product

Resources

About