Extreme learning machines: new trends and applications

Abstract

Extreme learning machine (ELM), as a new learning framework, draws increasing attractions in the areas of large-scale computing, high-speed signal processing, artificial intelligence, and so on. ELM aims to break the barriers between the conventional artificial learning techniques and biological learning mechanism and represents a suite of machine learning techniques in which hidden neurons need not to be tuned. ELM theories and algorithms argue that "random hidden neurons" capture the essence of some brain learning mechanisms as well as the intuitive sense that the efficiency of brain learning need not rely on computing power of neurons. Thus, compared with traditional neural networks and support vector machine, ELM offers significant advantages such as fast learning speed, ease of implementation, and minimal human intervention. Due to its remarkable generalization performance and implementation efficiency, ELM has been applied in various applications. In this paper, we first provide an overview of newly derived ELM theories and approaches. On the other hand, with the ongoing development of multilayer feature representation, some new trends on ELM-based hierarchical learning are discussed. Moreover, we also present several interesting ELM applications to showcase the practical advances on this subject.

Spatial Prediction of Soil Organic Matter Using a Hybrid Geostatistical Model of an Extreme Learning Machine and Ordinary Kriging

“…The ELM algorithm has gradually been employed to determine soil properties such as soil heavy metals, soil temperature, and soil moisture, due to its excellent generalization performance and global optimal property [34][35][36]. Most prominently, ELM has a faster learning speed than classical artificial neural networks (ANN) because it simplifies the training processes by randomly selecting the parameters.…”

Fast Learning for Big Data Using Dynamic Function

“…We use Matlab software running on Windows 8 machine with Intel Core i7 processor. The weights were generated randomly between [-0.4, 0.6] using equations (1)(2)(3)(4) The experiments were run 10 times for each value of error threshold and the average value is taken. The best results is shown in Table 1.…”

“…Extreme learning machine (ELM) [1] is one of the leading trends for fast learning. Unlike the other traditional learning algorithms, for example, Back Propagation-based neural networks, or support vector machine (SVM)], the parameters of hidden layers of ELM are randomly established and need not be tuned, thus the training of hidden nodes can be established before the inputs are acquired.…”

Haptic Identification by ELM-Controlled Uncertain Manipulator

“…Then, from the second trial, we update the reference point x * d and feedforward force f d in the reference model iteratively (39) and (40). We set 20 trials for exploration, and the estimations of K 0 and x b of the spring are shown in Fig.…”

“…Extreme learning machine (ELM) is a unified framework of the generalized single-hidden layer feedforward networks (SLFN) [38] [39]. It is distinctive that the weights of input samples and the bias of activation functions can be randomly chosen in the training process of the ELM.…”