A Fast Reduced Kernel Extreme Learning Machine

Deng, Wu; Ong, Yew-Soon; Zheng, Qinghua

doi:10.1016/j.neunet.2015.10.006

Cited by 70 publications

(32 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In ELM, the kernel is defined as elm = 푇 , and, according to Huang et al 's work in [37], it improves the generalization performance. In recent years, it is also used to speed up the training process [38].…”

Section: Extreme Learning Machinementioning

confidence: 99%

See 1 more Smart Citation

Weighted Domain Transfer Extreme Learning Machine and Its Online Version for Gas Sensor Drift Compensation in E‐Nose Systems

Ma¹,

Luo²,

Qin³

et al. 2018

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

Machine learning approaches have been widely used to tackle the problem of sensor array drift in E-Nose systems. However, labeled data are rare in practice, which makes supervised learning methods hard to be applied. Meanwhile, current solutions require updating the analytical model in an offline manner, which hampers their uses for online scenarios. In this paper, we extended Target Domain Adaptation Extreme Learning Machine (DAELM T) to achieve high accuracy with less labeled samples by proposing a Weighted Domain Transfer Extreme Learning Machine, which uses clustering information as prior knowledge to help select proper labeled samples and calculate sensitive matrix for weighted learning. Furthermore, we converted DAELM T and the proposed method into their online learning versions under which scenario the labeled data are selected beforehand. Experimental results show that, for batch learning version, the proposed method uses around 20% less labeled samples while achieving approximately equivalent or better accuracy. As for the online versions, the methods maintain almost the same accuracies as their offline counterparts do, but the time cost remains around a constant value while that of offline versions grows with the number of samples.

show abstract

Section: Extreme Learning Machinementioning

confidence: 99%

“…According to the work in [38,52], a special RBF function, that is, Gaussian RBF function, ensures that the inverses of 푇 and 푇 exist. Therefore, we use this type of function in the paper.…”

Section: Inputmentioning

confidence: 99%

Weighted Domain Transfer Extreme Learning Machine and Its Online Version for Gas Sensor Drift Compensation in E‐Nose Systems

Ma¹,

Luo²,

Qin³

et al. 2018

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

show abstract

“…In this section, an overview of ELM and its kernel extension versions: KELM [21] and RKELM [22] is presented. This serves to lay the foundation of the experiments in Sect.…”

Section: Extreme Learning Machine and Its Kernel Extension Versionsmentioning

confidence: 99%

“…To address the issue mentioned in the previous section, based on kernel theory and ELM, [22] presented a fast noniterative kernel machine, which is referred to as RKELM. The key characteristic of the RKELM algorithm is that support vectors are randomly chosen from the train set as opposed to some sophisticated process which is often compute intensive.…”

Section: Rkelmmentioning

confidence: 99%

“…However, the conventional SVM as well as the KELM do not scale well with the big datasets in general. For this problem, Deng et al [22] presented a fast and accurate kernel-based supervised algorithm referred to as the reduced kernel extreme learning machine (RKELM) which randomly selects a subset of the available data samples as support vectors (or mapping samples) instead of identifying the support vectors (or weight vectors) iteratively with respect to SVM.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An Extreme Learning Machine Architecture Based on Volterra Filtering and PCA

Chen

Yang

et al. 2017

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYExtreme learning machine (ELM) has recently attracted many researchers' interest due to its very fast learning speed, good generalization ability, and ease of implementation. However, it has a linear output layer which may limit the capability of exploring the available information, since higher-order statistics of the signals are not taken into account. To address this, we propose a novel ELM architecture in which the linear output layer is replaced by a Volterra filter structure. Additionally, the principal component analysis (PCA) technique is used to reduce the number of effective signals transmitted to the output layer. This idea not only improves the processing capability of the network, but also preserves the simplicity of the training process. Then we carry out performance evaluation and application analysis for the proposed architecture in the context of supervised classification and unsupervised equalization respectively, and the obtained results either on publicly available datasets or various channels, when compared to those produced by already proposed ELM versions and a state-of-the-art algorithm: support vector machine (SVM), highlight the adequacy and the advantages of the proposed architecture and characterize it as a promising tool to deal with signal processing tasks.

show abstract

Islanding and power quality disturbance monitoring in microgrid using adaptive cross variational mode decomposition and reduced kernel ridge regression

Dash

Satapathy

Nayak

2020

Int Trans Electr Energ Syst

View full text Add to dashboard Cite

Summary This article presents the detection and classification of islanding and power quality (PQ) disturbances for a multiple distributed generation based using an adaptive cross variational mode decomposition (XVMD) with reduced kernel ridge regression (RKRR). This article considers photovoltaic as the primary DG and studied the effect of the solar irradiation variation. Further the considered microgrid is subjected to different operating conditions pertaining to both islanding and non‐islanding disturbances. The proposed XVMD is an extension of VMD and with RKRR, it is a new contribution to the earlier studies to deal dynamic detection threshold requirement. The XVMD based RKRR needs simple features to classify the disturbances in less time as compared to the VMD based methods. The features to classify the events are calculated from the optimal mode selected by the hybrid firefly algorithm. For the performance analysis, voltage signals across the PV based DGs are obtained by initiating various power qualities and islanding disturbances. A validation case study through TMS 320C6713 Starter Kit (DSK) is presented for capacitor switching. The robustness of the proposed XVMD‐RKRR is observed from a large number of numerical experimentation by comparing it with other well‐known approaches.

show abstract

A Fast Reduced Kernel Extreme Learning Machine

Cited by 70 publications

References 19 publications

Weighted Domain Transfer Extreme Learning Machine and Its Online Version for Gas Sensor Drift Compensation in E‐Nose Systems

Weighted Domain Transfer Extreme Learning Machine and Its Online Version for Gas Sensor Drift Compensation in E‐Nose Systems

An Extreme Learning Machine Architecture Based on Volterra Filtering and PCA

Islanding and power quality disturbance monitoring in microgrid using adaptive cross variational mode decomposition and reduced kernel ridge regression

Contact Info

Product

Resources

About