Multilayer Extreme Learning Machine With Subnetwork Nodes for Representation Learning

Randomized algorithms for large-scale dictionary learning

Dictionary learning is an important sparse representation algorithm which has been widely used in machine learning and artificial intelligence. However, for massive data in the big data era, classical dictionary learning algorithms are computationally expensive and even can be infeasible. To overcome this difficulty, we propose new dictionary learning methods based on randomized algorithms. The contributions of this work are as follows. First, we find that dictionary matrix is often numerically low-rank. Based on this property, we apply randomized singular value decomposition (RSVD) to the dictionary matrix, and propose a randomized algorithm for linear dictionary learning. Compared with the classical K-SVD algorithm, an advantage is that one can update all the elements of the dictionary matrix simultaneously. Second, to the best of our knowledge, there are few theoretical results on why one can solve the involved matrix computation problems inexactly in dictionary learning. To fill-in this gap, we show the rationality of this randomized algorithm with inexact solving, from a matrix perturbation analysis point of view. Third, based on the numerically low-rank property and Nyström approximation of the kernel matrix, we propose a randomized kernel dictionary learning algorithm, and establish the distance between the exact solution and the computed solution, to show the effectiveness of the proposed randomized kernel dictionary learning algorithm. Fourth, we propose an efficient scheme for the testing stage in kernel dictionary learning. By using this strategy, there is no need to form nor store kernel matrices explicitly both in the training and the testing stages. Comprehensive numerical experiments are performed on some real-world data sets. Numerical results demonstrate the rationality of our strategies, and show that the proposed algorithms are much efficient than some state-of-the-art dictionary learning algorithms. The MATLAB codes of the proposed algorithms are publicly available from https://github.com/Jiali-yang/RALDL_RAKDL.

Semisupervised Manifold Regularization via a Subnetwork-Based Representation Learning Model

Semisupervised classification with a few labeled training samples is a challenging task in the area of data mining. Moore-Penrose inverse (MPI)-based manifold regularization (MR) is a widely used technique in tackling semisupervised classification. However, most of the existing MPI-based MR algorithms can only generate loosely connected feature encoding, which is generally less effective in data representation and feature learning. To alleviate this deficiency, we introduce a new semisupervised multilayer subnet neural network called SS-MSNN. The key contributions of this article are as follows: 1) a novel MPI-based MR model using the subnetwork structure is introduced. The subnet model is utilized to enrich the latent space representations iteratively; 2) a one-step training process to learn the discriminative encoding is proposed. The proposed SS-MSNN learns parameters by directly optimizing the entire network, accepting input from one end, and producing output at the other end; and 3) a new semisupervised dataset called HFSWR-RDE is built for this research. Experimental results on multiple domains show that the SS-MSNN achieves promising performance over the other semisupervised learning algorithms, demonstrating fast inference speed and better generalization ability.

D-BIN: A Generalized Disentangling Batch Instance Normalization for Domain Adaptation

Pattern recognition is significantly challenging in real-world scenarios by the variability of visual statistics. Therefore, most existing algorithms relying on the independent identically distributed assumption of training and test data suffer from the poor generalization capability of inference on unseen testing datasets. Although numerous studies, including domain discriminator or domain-invariant feature learning, are proposed to alleviate this problem, the data-driven property and lack of interpretation of their principle throw researchers and developers off. Consequently, this dilemma incurs us to rethink the essence of networks' generalization. An observation that visual patterns cannot be discriminative after style transfer inspires us to take careful consideration of the importance of style features and content features. Does the style information related to the domain bias? How to effectively disentangle content and style features across domains? In this article, we first investigate the effect of feature normalization on domain adaptation. Based on it, we propose a novel normalization module to adaptively leverage the propagated information through each channel and batch of features called disentangling batch instance normalization (D-BIN). In this module, we explicitly explore domain-specific and domaininvariant feature disentanglement. We maneuver contrastive learning to encourage images with the same semantics from different domains to have similar content representations while having dissimilar style representations. Furthermore, we construct both self-form and dual-form regularizers for preserving the mutual information (MI) between feature representations of the normalization layer in order to compensate for the loss of discriminative information and effectively match the distributions across domains. D-BIN and the constrained term can be simply plugged into state-of-the-art (SOTA) networks to improve their performance. In the end, experiments, including domain adaptation and generalization, conducted on different datasets have proven their effectiveness.

Sequential Order-Aware Coding-Based Robust Subspace Clustering for Human Action Recognition in Untrimmed Videos

Human action recognition (HAR) is one of most important tasks in video analysis. Since video clips distributed on networks are usually untrimmed, it is required to accurately segment a given untrimmed video into a set of action segments for HAR. As an unsupervised temporal segmentation technology, subspace clustering learns the codes from each video to construct an affinity graph, and then cuts the affinity graph to cluster the video into a set of action segments. However, most of the existing subspace clustering schemes not only ignore the sequential information of frames in code learning, but also the negative effects of noises when cutting the affinity graph, which lead to inferior performance. To address these issues, we propose a sequential order-aware coding-based robust subspace clustering (SOAC-RSC) scheme for HAR. By feeding the motion features of video frames into multi-layer neural networks, two expressive code matrices are learned in a sequential order-aware manner from unconstrained and constrained videos, respectively, to construct the corresponding affinity graphs. Then, with the consideration of the existence of noise effects, a simple yet robust cutting algorithm is proposed to cut the constructed affinity graphs to accurately obtain the action segments for HAR. The extensive experiments demonstrate the proposed SOAC-RSC scheme achieves the state-of-the-art performance on the datasets of Keck Gesture and Weizmann, and provides competitive performance on the other 6 public datasets such as UCF101 and URADL for HAR task, compared to the recent related approaches.

Human emotion recognition from EEG-based brain–computer interface using machine learning: a comprehensive review

Affective computing, a subcategory of artificial intelligence, detects, processes, interprets, and mimics human emotions. Thanks to the continued advancement of portable non-invasive human sensor technologies, like brain–computer interfaces (BCI), emotion recognition has piqued the interest of academics from a variety of domains. Facial expressions, speech, behavior (gesture/posture), and physiological signals can all be used to identify human emotions. However, the first three may be ineffectual because people may hide their true emotions consciously or unconsciously (so-called social masking). Physiological signals can provide more accurate and objective emotion recognition. Electroencephalogram (EEG) signals respond in real time and are more sensitive to changes in affective states than peripheral neurophysiological signals. Thus, EEG signals can reveal important features of emotional states. Recently, several EEG-based BCI emotion recognition techniques have been developed. In addition, rapid advances in machine and deep learning have enabled machines or computers to understand, recognize, and analyze emotions. This study reviews emotion recognition methods that rely on multi-channel EEG signal-based BCIs and provides an overview of what has been accomplished in this area. It also provides an overview of the datasets and methods used to elicit emotional states. According to the usual emotional recognition pathway, we review various EEG feature extraction, feature selection/reduction, machine learning methods (e.g., k-nearest neighbor), support vector machine, decision tree, artificial neural network, random forest, and naive Bayes) and deep learning methods (e.g., convolutional and recurrent neural networks with long short term memory). In addition, EEG rhythms that are strongly linked to emotions as well as the relationship between distinct brain areas and emotions are discussed. We also discuss several human emotion recognition studies, published between 2015 and 2021, that use EEG data and compare different machine and deep learning algorithms. Finally, this review suggests several challenges and future research directions in the recognition and classification of human emotional states using EEG.

Multimodal Vigilance Estimation Using Deep Learning

The phenomenon of increasing accidents caused by reduced vigilance does exist. In the future, the high accuracy of vigilance estimation will play a significant role in public transportation safety. We propose a multimodal regression network that consists of multichannel deep autoencoders with subnetwork neurons (MCDAE _sn ). After we define two thresholds of "0.35" and "0.70" from the percentage of eye closure, the output values are in the continuous range of 0-0.35, 0.36-0.70, and 0.71-1 representing the awake state, the tired state, and the drowsy state, respectively. To verify the efficiency of our strategy, we first applied the proposed approach to a single modality. Then, for the multimodality, since the complementary information between forehead electrooculography and electroencephalography features, we found the performance of the proposed approach using features fusion significantly improved, demonstrating the effectiveness and efficiency of our method.

Multimodel Feature Reinforcement Framework Using Moore-Penrose Inverse for Big Data Analysis

Fully connected representation learning (FCRL) is one of the widely used network structures in multimodel image classification frameworks. However, most FCRL-based structures, for instance, stacked autoencoder encode features and find the final cognition with separate building blocks, resulting in loosely connected feature representation. This article achieves a robust representation by considering a low-dimensional feature and the classifier model simultaneously. Thus, a new hierarchical subnetwork-based neural network (HSNN) is proposed in this article. The novelties of this framework are as follows: 1) it is an iterative learning process, instead of stacking separate blocks to obtain the discriminative encoding and the final classification results. In this sense, the optimal global features are generated; 2) it applies Moore-Penrose (MP) inverse-based batch-by-batch learning strategy to handle large-scale data sets, so that large data set, such as Place365 containing 1.8 million images, can be processed effectively. The experimental results on multiple domains with a varying number of training samples from  ∼  1 K to  ∼ 2 M show that the proposed feature reinforcement framework achieves better generalization performance compared with most state-of-the-art FCRL methods.

Deep Learning-Based Model Predictive Control for Continuous Stirred-Tank Reactor System

A continuous stirred-tank reactor (CSTR) system is widely applied in wastewater treatment processes. Its control is a challenging industrial-process-control problem due to great difficulty to achieve accurate system identification. This work proposes a deep learning-based model predictive control (DeepMPC) to model and control the CSTR system. The proposed DeepMPC consists of a growing deep belief network (GDBN) and an optimal controller. First, GDBN can automatically determine its size with transfer learning to achieve high performance in system identification, and it serves just as a predictive model of a controlled system. The model can accurately approximate the dynamics of the controlled system with a uniformly ultimately bounded error. Second, quadratic optimization is conducted to obtain an optimal controller. This work analyzes the convergence and stability of DeepMPC. Finally, the DeepMPC is used to model and control a second-order CSTR system. In the experiments, DeepMPC shows a better performance in modeling, tracking, and antidisturbance than the other state-of-the-art methods.

Recomputation of the Dense Layers for Performance Improvement of DCNN

Gradient descent optimization of learning has become a paradigm for training deep convolutional neural networks (DCNN). However, utilizing other learning strategies in the training process of the DCNN has rarely been explored by the deep learning (DL) community. This serves as the motivation to introduce a non-iterative learning strategy to retrain neurons at the top dense or fully connected (FC) layers of DCNN, resulting in, higher performance. The proposed method exploits the Moore-Penrose Inverse to pull back the current residual error to each FC layer, generating well-generalized features. Further, the weights of each FC layers are recomputed according to the Moore-Penrose Inverse. We evaluate the proposed approach on six most widely accepted object recognition benchmark datasets: Scene-15, CIFAR-10, CIFAR-100, SUN-397, Places365, and ImageNet. The experimental results show that the proposed method obtains improvements over 30 state-of-the-art methods. Interestingly, it also indicates that any DCNN with the proposed method can provide better performance than the same network with its original Backpropagation (BP)-based training.

Human motion recognition based on limit learning machine

With the development of technology, human motion capture data have been widely used in the fields of human–computer interaction, interactive entertainment, education, and medical treatment. As a problem in the field of computer vision, human motion recognition has become a key technology in somatosensory games, security protection, and multimedia information retrieval. Therefore, it is important to improve the recognition rate of human motion. Based on the above background, the purpose of this article is human motion recognition based on extreme learning machine. Based on the existing action feature descriptors, this article makes improvements to features and classifiers and performs experiments on the Microsoft model specific register (MSR)-Action3D data set and the Bonn University high density metal (HDM05) motion capture data set. Based on displacement covariance descriptor and direction histogram descriptor, this article described both combine to produce a new combination; the description can statically reflect the joint position relevant information and at the same time, the change information dynamically reflects the joint position, uses the extreme learning machine for classification, and gets better recognition result. The experimental results show that the combined descriptor and extreme learning machine recognition rate on these two data sets is significantly improved by about 3% compared with the existing methods.

Features Combined From Hundreds of Midlayers: Hierarchical Networks With Subnetwork Nodes

In this paper, we believe that the mixed selectivity of neuron in the top layer encodes distributed information produced from other neurons to offer a significant computational advantage over recognition accuracy. Thus, this paper proposes a hierarchical network framework that the learning behaviors of features combined from hundreds of midlayers. First, a subnetwork neuron, which itself could be constructed by other nodes, is functional as a subspace features extractor. The top layer of a hierarchical network needs subspace features produced by the subnetwork neurons to get rid of factors that are not relevant, but at the same time, to recast the subspace features into a mapping space so that the hierarchical network can be processed to generate more reliable cognition. Second, this paper shows that with noniterative learning strategy, the proposed method has a wider and shallower structure, providing a significant role in generalization performance improvements. Hence, compared with other state-of-the-art methods, multiple channel features with the proposed method could provide a comparable or even better performance, which dramatically boosts the learning speed. Our experimental results show that our platform can provide a much better generalization performance than 55 other state-of-the-art methods.

Multi-View Fusion with Extreme Learning Machine for Clustering

Unlabeled, multi-view data presents a considerable challenge in many real-world data analysis tasks. These data are worth exploring because they often contain complementary information that improves the quality of the analysis results. Clustering with multi-view data is a particularly challenging problem as revealing the complex data structures between many feature spaces demands discriminative features that are specific to the task and, when too few of these features are present, performance suffers. Extreme learning machines (ELMs) are an emerging form of learning model that have shown an outstanding representation ability and superior performance in a range of different learning tasks. Motivated by the promise of this advancement, we have developed a novel multi-view fusion clustering framework based on an ELM, called MVEC. MVEC learns the embeddings from each view of the data via the ELM network, then constructs a single unified embedding according to the correlations and dependencies between each embedding and automatically weighting the contribution of each. This process exposes the underlying clustering structures embedded within multi-view data with a high degree of accuracy. A simple yet efficient solution is also provided to solve the optimization problem within MVEC. Experiments and comparisons on eight different benchmarks from different domains confirm MVEC’s clustering accuracy.

Cholesky Factorization Based Online Sequential Extreme Learning Machines with Persistent Regularization and Forgetting Factor

The online sequential extreme learning machine with persistent regularization and forgetting factor (OSELM-PRFF) can avoid potential singularities or ill-posed problems of online sequential regularized extreme learning machines with forgetting factors (FR-OSELM), and is particularly suitable for modelling in non-stationary environments. However, existing algorithms for OSELM-PRFF are time-consuming or unstable in certain paradigms or parameters setups. This paper presents a novel algorithm for OSELM-PRFF, named “Cholesky factorization based” OSELM-PRFF (CF-OSELM-PRFF), which recurrently constructs an equation for extreme learning machine and efficiently solves the equation via Cholesky factorization during every cycle. CF-OSELM-PRFF deals with timeliness of samples by forgetting factor, and the regularization term in its cost function works persistently. CF-OSELM-PRFF can learn data one-by-one or chunk-by-chunk with a fixed or varying chunk size. Detailed performance comparisons between CF-OSELM-PRFF and relevant approaches are carried out on several regression problems. The numerical simulation results show that CF-OSELM-PRFF demonstrates higher computational efficiency than its counterparts, and can yield stable predictions.

Taste Recognition in E-Tongue Using Local Discriminant Preservation Projection

Electronic tongue (E-Tongue), as a novel taste analysis tool, shows a promising perspective for taste recognition. In this paper, we constructed a voltammetric E-Tongue system and measured 13 different kinds of liquid samples, such as tea, wine, beverage, functional materials, etc. Owing to the noise of system and a variety of environmental conditions, the acquired E-Tongue data shows inseparable patterns. To this end, from the viewpoint of algorithm, we propose a local discriminant preservation projection (LDPP) model, an under-studied subspace learning algorithm, that concerns the local discrimination and neighborhood structure preservation. In contrast with other conventional subspace projection methods, LDPP has two merits. On one hand, with local discrimination it has a higher tolerance to abnormal data or outliers. On the other hand, it can project the data to a more separable space with local structure preservation. Further, support vector machine, extreme learning machine (ELM), and kernelized ELM (KELM) have been used as classifiers for taste recognition in E-Tongue. Experimental results demonstrate that the proposed E-Tongue is effective for multiple tastes recognition in both efficiency and effectiveness. Particularly, the proposed LDPP-based KELM classifier model achieves the best taste recognition performance of 98%. The developed benchmark data sets and codes will be released and downloaded in http://www.leizhang.tk/ tempcode.html.

Robust Learning With Kernel Mean -Power Error Loss

Correntropy is a second order statistical measure in kernel space, which has been successfully applied in robust learning and signal processing. In this paper, we define a nonsecond order statistical measure in kernel space, called the kernel mean- power error (KMPE), including the correntropic loss (C-Loss) as a special case. Some basic properties of KMPE are presented. In particular, we apply the KMPE to extreme learning machine (ELM) and principal component analysis (PCA), and develop two robust learning algorithms, namely ELM-KMPE and PCA-KMPE. Experimental results on synthetic and benchmark data show that the developed algorithms can achieve better performance when compared with some existing methods.

Learning a single-hidden layer feedforward neural network using a rank correlation-based strategy with application to high dimensional gene expression and proteomic spectra datasets in cancer detection

Methods based on microarrays (MA), mass spectrometry (MS), and machine learning (ML) algorithms have evolved rapidly in recent years, allowing for early detection of several types of cancer. A pitfall of these approaches, however, is the overfitting of data due to large number of attributes and small number of instances -- a phenomenon known as the 'curse of dimensionality'. A potentially fruitful idea to avoid this drawback is to develop algorithms that combine fast computation with a filtering module for the attributes. The goal of this paper is to propose a statistical strategy to initiate the hidden nodes of a single-hidden layer feedforward neural network (SLFN) by using both the knowledge embedded in data and a filtering mechanism for attribute relevance. In order to attest its feasibility, the proposed model has been tested on five publicly available high-dimensional datasets: breast, lung, colon, and ovarian cancer regarding gene expression and proteomic spectra provided by cDNA arrays, DNA microarray, and MS. The novel algorithm, called adaptive SLFN (aSLFN), has been compared with four major classification algorithms: traditional ELM, radial basis function network (RBF), single-hidden layer feedforward neural network trained by backpropagation algorithm (BP-SLFN), and support vector-machine (SVM). Experimental results showed that the classification performance of aSLFN is competitive with the comparison models.

A Novel AdaBoost Framework With Robust Threshold and Structural Optimization

The AdaBoost algorithm is a popular ensemble method that combines several weak learners to boost generalization performance. However, conventional AdaBoost.RT algorithms suffer from the limitation that the threshold value must be manually specified rather than chosen through a self-adaptive mechanism, which cannot guarantee a result in an optimal model for general cases. In this paper, we present a generic AdaBoost framework with robust threshold mechanism and structural optimization on regression problems. The error statistics of each weak learner on one given problem dataset is utilized to automate the choice of the optimal cut-off threshold value. In addition, a special single-layer neural network is employed to provide a second opportunity to further adjust the structure and strength the adaption capability of the AdaBoost regression model. Moreover, to consolidate the theoretical foundation of AdaBoost algorithms, we are the first to conduct a rigorous and comprehensive theoretical analysis on the proposed approach. We prove that the general bound on the empirical error with a fraction of training examples is always within a limited soft margin, which indicates that our novel algorithm can avoid over-fitting. We further analyze the bounds on the generalization error directly under probably approximately correct learning. The extensive experimental verifications on the UCI benchmarks have demonstrated that the performance of the proposed method is superior to other state-of-the-art ensemble and single learning algorithms. Furthermore, a real-world indoor positioning application has also revealed that the proposed method has higher positioning accuracy and faster speed.

An Extreme Learning Machine Approach to Density Estimation Problems

In this paper, we discuss how the extreme learning machine (ELM) framework can be effectively employed in the unsupervised context of multivariate density estimation. In particular, two algorithms are introduced, one for the estimation of the cumulative distribution function underlying the observed data, and one for the estimation of the probability density function. The algorithms rely on the concept of F -discrepancy, which is closely related to the Kolmogorov-Smirnov criterion for goodness of fit. Both methods retain the key feature of the ELM of providing the solution through random assignment of the hidden feature map and a very light computational burden. A theoretical analysis is provided, discussing convergence under proper hypotheses on the chosen activation functions. Simulation tests show how ELMs can be successfully employed in the density estimation framework, as a possible alternative to other standard methods.

Urban Traffic Congestion Evaluation Based on Kernel the Semi-Supervised Extreme Learning Machine

There is always an asymmetric phenomenon between traffic data quantity and unit information content. Labeled data is more effective but scarce, while unlabeled data is large but weaker in sample information. In an urban transportation assessment system, semi-supervised extreme learning machine (SSELM) can unite manual observed data and extensively collected data cooperatively to build connection between congestion condition and road information. In our method, semi-supervised learning can integrate both small-scale labeled data and large-scale unlabeled data, so that they can play their respective advantages, while the ELM can process large scale data at high speed. Optimized by kernel function, Kernel-SSELM can achieve higher classification accuracy and robustness than original SSELM. Both the experiment and the real-time application show that the evaluation system can precisely reflect the traffic condition.