Decades of research have shown machine learning superiority in discovering highly nonlinear patterns embedded in electroencephalography (EEG) records compared with conventional statistical techniques. However, even the most advanced machine learning techniques require relatively large, labeled EEG repositories. EEG data collection and labeling are costly. Moreover, combining available datasets to achieve a large data volume is usually infeasible due to inconsistent experimental paradigms across trials. Self-supervised learning (SSL) solves these challenges because it enables learning from EEG records across trials with variable experimental paradigms, even when the trials explore different phenomena. It aggregates multiple EEG repositories to increase accuracy, reduce bias, and mitigate overfitting in machine learning training. In addition, SSL could be employed in situations where there is limited labeled training data, and manual labeling is costly. This article: 1) provides a brief introduction to SSL; 2) describes some SSL techniques employed in recent studies, including EEG; 3) proposes current and potential SSL techniques for future investigations in EEG studies; 4) discusses the cons and pros of different SSL techniques; and 5) proposes holistic implementation tips and potential future directions for EEG SSL practices.

The present paper introduces a novel method, named AutoEncoder-Filter Bank Common Spatial Patterns (AE-FBCSP), to decode imagined movements from electroencephalography (EEG). AE-FBCSP is an extension of the well-established FBCSP and is based on a global (cross-subject) and subsequent transfer learning subject-specific (intra-subject) approach. A multi-way extension of AE-FBCSP is also introduced in this paper. Features are extracted from high-density EEG (64 electrodes), by means of FBCSP, and used to train a custom AE, in an unsupervised way, to project the features into a compressed latent space. Latent features are used to train a supervised classifier (feed forward neural network) to decode the imagined movement. The proposed method was tested using a public dataset of EEGs collected from 109 subjects. The dataset consists of right-hand, left-hand, both hands, both feet motor imagery and resting EEGs. AE-FBCSP was extensively tested in the 3-way classification (right hand vs left hand vs resting) and also in the 2-way, 4-way and 5-way ones, both in cross- and intra-subject analysis. AE-FBCSP outperformed standard FBCSP in a statistically significant way (p > 0.05) and achieved a subject-specific average accuracy of 89.09% in the 3-way classification. The proposed methodology performed subject-specific classification better than other comparable methods in the literature, applied to the same dataset, also in the 2-way, 4-way and 5-way tasks. One of the most interesting outcomes is that AE-FBCSP remarkably increased the number of subjects that responded with a very high accuracy, which is a fundamental requirement for BCI systems to be applied in practice.

Abstract Cities that are adopting innovative and technology‐driven solutions to improve the city's efficiency are consideredsmart cities. With the increased attention on smart cities with self‐driving vehicles, drones, and robots, designing smart infrastructure is only a natural extension. Smart infrastructures aim to self‐diagnose, self‐power, self‐adapt, and self‐heal during normal and extreme operating conditions. Structural vibration control (SVC) and structural health monitoring (SHM) technologies, in particular, are expected to play pivotal roles in the development of modern smart and resilient structures. SVC methodologies intend to provide supplemental damping and reduce the structural dynamic responses during normal and extreme events. SHM methodologies offer valuable information about the structure's condition that is useful for maintenance purposes and rapid damage detection in post‐hazard events. The collapse of the 12‐story Champlain Towers South, a beachfront condominium in the Miami suburb of Surfside, Florida, could have been known in advance with an embedded SHM technology. More recently, the integrated structural control and health monitoring (ISCHM) systems have shown promise in the development of smart cities of the future. The integrated architecture incorporates the control and health monitoring components as complementary technologies and simultaneously takes advantage of both technologies. This article provides a state‐of‐the‐art review of ISCHM ideas and systems. It presents recent significant developments in structural control, SHM, and energy harvesting that are paving the way towards the advent of integrated ISCHM systems, including damage‐tolerant control systems. This article also identifies future promising research areas for designing the next generation of autonomous ISCHM systems for smart cities.

Abstract In the Big Data Era, recommender systems perform a fundamental role in data management and information filtering. In this context, Collaborative Filtering (CF) persists as one of the most prominent strategies to effectively deal with large datasets and is capable of offering users interesting content in a recommendation fashion. Nevertheless, it is well‐known CF recommenders suffer from data sparsity, mainly in cold‐start scenarios, substantially reducing the quality of recommendations. In the vast literature about the aforementioned topic, there are numerous solutions, in which the state‐of‐the‐art contributions are, in some sense, conditioned or associated with traditional CF methods such as Matrix Factorization (MF), that is, they rely on linear optimization procedures to model users and items into low‐dimensional embeddings. To overcome the aforementioned challenges, there has been an increasing number of studies exploring deep learning techniques in the CF context for latent factor modelling. In this research, authors conduct a systematic review focusing on state‐of‐the‐art literature on deep learning techniques applied in collaborative filtering recommendation, and also featuring primary studies related to mitigating the cold start problem. Additionally, authors considered the diverse non‐linear modelling strategies to deal with rating data and side information, the combination of deep learning techniques with traditional CF‐based linear methods, and an overview of the most used public datasets and evaluation metrics concerning CF scenarios.

This paper presents a novel dynamic ensemble learning (DEL) algorithm for designing ensemble of neural networks (NNs). DEL algorithm determines the size of ensemble, the number of individual NNs employing a constructive strategy, the number of hidden nodes of individual NNs employing a constructive–pruning strategy, and different training samples for individual NN’s learning. For diversity, negative correlation learning has been introduced and also variation of training samples has been made for individual NNs that provide better learning from the whole training samples. The major benefits of the proposed DEL compared to existing ensemble algorithms are (1) automatic design of ensemble; (2) maintaining accuracy and diversity of NNs at the same time; and (3) minimum number of parameters to be defined by user. DEL algorithm is applied to a set of real-world classification problems such as the cancer, diabetes, heart disease, thyroid, credit card, glass, gene, horse, letter recognition, mushroom, and soybean datasets. It has been confirmed by experimental results that DEL produces dynamic NN ensembles of appropriate architecture and diversity that demonstrate good generalization ability.

Abstract An important subfield of brain–computer interface is the classification of motor imagery (MI) signals where a presumed action, for example, imagining the hands' motions, is mentally simulated. The brain dynamics of MI is usually measured by electroencephalography (EEG) due to its noninvasiveness. The next generation of brain–computer interface systems can benefit from the generative deep learning (GDL) models by providing end‐to‐end (e2e) machine learning and increasing their accuracy. In this study, to exploit the e2e‐property of deep learning models, a novel GDL methodology is proposed where only minimal objective‐free preprocessing steps are needed. Furthermore, to deal with the complicated multi‐class MI–EEG signals, an innovative multilevel GDL‐based classifying scheme is proposed. The effectiveness of the proposed model and its robustness against noisy MI–EEG signals is evaluated using two different GDL models, that is, deep belief network and stacked sparse autoencoder in e2e manner. Experimental results demonstrate the effectiveness of the proposed methodology with improved accuracy compared with the widely used filter bank common spatial patterns algorithm.

Abstract Artificial intelligence and expert system remains a key technology in the 21st century. Using active controllers, a structure can adaptively adjust its behaviour during dynamic loads. Such structures with self‐modifying capabilities are referred to as intelligent or smart structures. Smart structure technology has the potential to be a game changer in the structural engineering field. It promises to have enormous consequences in terms of preventing loss of life and damage to structure and their content especially for large structures with hundreds or thousands of components. A key element in successful implementation of smart active control technology is an effective control algorithm to compute the magnitudes of actual forces to be applied to the structure. In this paper, an overview of main active control methodologies for vibration control of smart civil and mechanical structures subjected to external dynamic loads is presented. The advantages and the disadvantages of different control algorithms are discussed. Finally, new trends in control algorithm research are pointed out including multiparadigm strategies, decentralized control, application of deep neural network machine learning techniques, control design for sustainability, and unification of the two fields of structural health monitoring and vibration control.

Abstract Compaction quality assessment and control for an earth‐rock dam is the key measure to ensure dam safety. However, to date, the compaction quality assessment model has not been accurate enough, and no effective feedback control measures have been developed. Hybrid data mining algorithms have great potential for solving this problem. In this study, smart bacteria‐foraging algorithm‐based customized kernel support vector regression (SBFA‐CKSVR) is proposed for compaction quality assessment, whereas an enhanced probabilistic neural network (EPNN) is adopted for compaction quality control. SBFA integrates a bacteria‐foraging algorithm, chaos mapping, and adaptive and quantum computing to solve the high‐dimensional complex problem effectively. CKSVR is proposed to approximate a function in quadratic continuous integral space L2(R) where its hyperparameters are optimized by SBFA. Finally, SBFA‐CKSVR is used to establish a high‐precision compaction quality assessment model whereas the EPNN is adopted to realize the compaction quality feedback control. A three‐dimensional real‐time monitoring system for the earth‐rock dam is also developed based on SBFA‐CKSVR and EPNN. A large‐scale hydraulic engineering application proves the effectiveness and superiority of this research compared with the previous work.

In this paper, a novel electroencephalographic (EEG)-based method is introduced for the quantification of brain-electrical connectivity changes over a longitudinal evaluation of mild cognitive impaired (MCI) subjects. In the proposed method, a dissimilarity matrix is constructed by estimating the coupling strength between every pair of EEG signals, Hierarchical clustering is then applied to group the related electrodes according to the dissimilarity estimated on pairs of EEG recordings. Subsequently, the connectivity density of the electrodes network is calculated. The technique was tested over two different coupling strength descriptors: wavelet coherence (WC) and permutation Jaccard distance (PJD), a novel metric of coupling strength between time series introduced in this paper. Twenty-five MCI patients were enrolled within a follow-up program that consisted of two successive evaluations, at time T0 and at time T1, three months later. At T1, four subjects were diagnosed to have converted to Alzheimer's Disease (AD). When applying the PJD-based method, the converted patients exhibited a significantly increased PJD (p < 0.05), i.e., a reduced overall coupling strength, specifically in delta and θ bands and in the overall range (0.5-32 Hz). In addition, in contrast to stable MCI patients, converted patients exhibited a network density reduction in every subband (delta, θ, alpha, and beta). When WC was used as coupling strength descriptor, the method resulted in a less sensitive and specific outcome. The proposed method, mixing nonlinear analysis to a machine learning approach, appears to provide an objective evaluation of the connectivity density modifications associated to the MCI-AD conversion, just processing noninvasive EEG signals.

Abstract Recently, multi‐ and many‐objective meta‐heuristic algorithms have received considerable attention due to their capability to solve optimization problems that require more than one fitness function. This paper presents a comprehensive study of these techniques applied in the context of machine learning problems. Three different topics are reviewed in this work: (a) feature extraction and selection, (b) hyper‐parameter optimization and model selection in the context of supervised learning, and (c) clustering or unsupervised learning. The survey also highlights future research towards related areas.

Abstract This paper presents a state‐of‐the‐art review of recent control algorithms used primarily for vibration control of smart structures and machines. Control algorithms can be divided into open loop, feedback, feedforward, and combined feedback and feedforward control. Solutions proposed can be subdivided into (a) classical control, (b) optimal control, (c) robust control, (d) intelligent control, and (e) self‐sustaining solutions and energy harvesting used in centralized and decentralized configurations. The most recent research on novel autonomous control systems intersects distributed control theory, agent‐based modeling, bioinspired computational learning, and optimization algorithms.

The keys for the development of an effective classification algorithm are: 1) discovering feature spaces with large margins between clusters and close proximity of the classmates and 2) discovering the smallest number of the features to perform accurate classification. In this paper, a new supervised classification algorithm, called neural dynamic classification (NDC), is presented with the goal of: 1) discovering the most effective feature spaces and 2) finding the optimum number of features required for accurate classification using the patented robust neural dynamic optimization model of Adeli and Park. The new classification algorithm is compared with the probabilistic neural network (PNN), enhanced PNN (EPNN), and support vector machine using two sets of classification problems. The first set consists of five standard benchmark problems. The second set is a large benchmark problem called Mixed National Institute of Standards and Technology database of handwritten digits. In general, NDC yields the most accurate classification results followed by EPNN. A beauty of the new algorithm is the smoothness of convergence curves which is an indication of robustness and good performance of the algorithm. The main aim is to maximize the prediction accuracy.

Alzheimer's disease (AD) is a degenerative neurological disorder characterized by a loss of functional connections between different areas of the brain. AD is considered a cortical dementia, thus Electroencephalography (EEG) has been used as a tool for diagnosing AD for the last two decades. Often, the hallmarks of EEG abnormality in AD patients are a shift of the power spectrum to lower frequencies and reduced coherences among cortical regions, however, it is still mostly unknown how these abnormalities evolve together with the disease progression. In this paper we proposed a longitudinal study of the EEG of three AD patients in order to study the disease progression, from the coherence point of view, over the four major EEG sub-bands: delta, theta, alpha and beta. The EEG was recorded at time T0 and then after three months (time T1). We proposed a coherence based hierarchical clustering method to group the electrodes together according to their mutual pairwise coherence, in order to evaluate how the brain connectivity changed along with the disease in the spectral domain. The results provide an in-depth view of the structure of electrode interconnection of every single patient in every sub-band at time T0 and time T1. This study endorsed the commonly shared belief that coherence reduces over time but it revealed that coherence spatial distribution changes in a different way, from patient to patient. The results also showed that a patient-specific brain connectivity analysis is possible and that a personalized analysis of the disease's progression might provide valuable diagnostic information. In the near future, the study will be extended to a larger dataset in order to validate the method statistically.

Abstract This paper presents a review of recently developed physics‐based search and optimization algorithms that have been inspired by natural phenomena. They include Big Bang–Big Crunch, black hole search, galaxy‐based search, artificial physics optimization, electromagnetism optimization, charged system search, colliding bodies optimization, and particle collision algorithm.

Gravitational search algorithm (GSA) is a nature-inspired conceptual framework with roots in gravitational kinematics, a branch of physics that models the motion of masses moving under the influence of gravity. In GSA, a collection of objects interacts with each other under the Newtonian gravity and the laws of motion. The performances of objects are measured by masses. All these objects attract each other by the gravity force, while this force causes a global movement of all objects toward the objects with heavier masses. The position of the object corresponds to a solution of the problem. The positions of the objects are updated every iteration and the best fitness along with its corresponding object is stored. Heavier masses move slowly than lighter ones. The algorithm terminates after a specified number of iterations after which the best fitness becomes the global fitness for a particular problem and the positions of the corresponding object becomes the global solution of that problem. This paper presents a review of GSA and its variants.

In the past three decades nature-inspired and meta-heuristic algorithms have dominated the literature in the broad areas of search and optimization. Harmony search algorithm (HSA) is a music-inspired population-based meta-heuristic search and optimization algorithm. The concept behind the algorithm is to find a perfect state of harmony determined by aesthetic estimation. This paper starts with an overview of the harmonic phenomenon in music and music improvisation used by musicians and how it is applied to the optimization problem. The concept of harmony memory and its mathematical implementation are introduced. A review of HSA and its variants is presented. Guidelines from the literature on the choice of parameters used in HSA for effective solution of optimization problems are summarized.

This paper proposes a novel method of corporate financial risk prediction (FRP) modeling called the adaptive and dynamic ensemble (ADE) of support vector machine (SVM) (ADE-SVM), which integrates the inflow of new data batches for FRP with the process of time. Namely, the characteristic change of corporate financial distress hidden in the data flow is considered as the concept drift of financial distress, and it is handled by ADE-SVM that keeps updating in time. Using the criteria of predictive ability and classifier diversity, the SVM ensemble is dynamically constructed by adaptively selecting the current base SVMs from candidate ones. The candidate SVMs are incrementally updated by considering the newest data batch at each new current time point. The results of the base SVMs are dynamically weighted by their validation accuracies on the latest data batch to generate the final prediction. Experiments were carried out on real-world data sets with current data for training and future data for testing. The results show that ADE-SVM overall outperforms the other three traditional dynamic modeling methods, particularly for harder FRP task with more insufficient information and more obvious concept drift.

Most current Artificial Neural Network (ANN) models are based on highly simplified brain dynamics. They have been used as powerful computational tools to solve complex pattern recognition, function estimation, and classification problems. ANNs have been evolving towards more powerful and more biologically realistic models. In the past decade, Spiking Neural Networks (SNNs) have been developed which comprise of spiking neurons. Information transfer in these neurons mimics the information transfer in biological neurons, i.e., via the precise timing of spikes or a sequence of spikes. To facilitate learning in such networks, new learning algorithms based on varying degrees of biological plausibility have also been developed recently. Addition of the temporal dimension for information encoding in SNNs yields new insight into the dynamics of the human brain and could result in compact representations of large neural networks. As such, SNNs have great potential for solving complicated time-dependent pattern recognition problems because of their inherent dynamic representation. This article presents a state-of-the-art review of the development of spiking neurons and SNNs, and provides insight into their evolution as the third generation neural networks.

Neural networks are investigated for predicting the magnitude of the largest seismic event in the following month based on the analysis of eight mathematically computed parameters known as seismicity indicators. The indicators are selected based on the Gutenberg-Richter and characteristic earthquake magnitude distribution and also on the conclusions drawn by recent earthquake prediction studies. Since there is no known established mathematical or even empirical relationship between these indicators and the location and magnitude of a succeeding earthquake in a particular time window, the problem is modeled using three different neural networks: a feed-forward Levenberg-Marquardt backpropagation (LMBP) neural network, a recurrent neural network, and a radial basis function (RBF) neural network. Prediction accuracies of the models are evaluated using four different statistical measures: the probability of detection, the false alarm ratio, the frequency bias, and the true skill score or R score. The models are trained and tested using data for two seismically different regions: Southern California and the San Francisco bay region. Overall the recurrent neural network model yields the best prediction accuracies compared with LMBP and RBF networks. While at the present earthquake prediction cannot be made with a high degree of certainty this research provides a scientific approach for evaluating the short-term seismic hazard potential of a region.

Two neural network models, called clustering-RBFNN and clustering-BPNN models, are created for estimating the work zone capacity in a freeway work zone as a function of seventeen different factors through judicious integration of the subtractive clustering approach with the radial basis function (RBF) and the backpropagation (BP) neural network models. The clustering-RBFNN model has the attractive characteristics of training stability, accuracy, and quick convergence. The results of validation indicate that the work zone capacity can be estimated by clustering-neural network models in general with an error of less than 10%, even with limited data available to train the models. The clustering-RBFNN model is used to study several main factors affecting work zone capacity. The results of such parametric studies can assist work zone engineers and highway agencies to create effective traffic management plans (TMP) for work zones quantitatively and objectively.

A new algorithm is presented for training of multilayer feedforward neural networks by integrating a genetic algorithm with an adaptive conjugate gradient neural network learning algorithm. The parallel hybrid learning algorithm has been implemented in C on an MIMD shared memory machine (Cray Y-MP8/864 supercomputer). It has been applied to two different domains, engineering design and image recognition. The performance of the algorithm has been evaluated by applying it to three examples. The superior convergence property of the parallel hybrid neural network learning algorithm presented in this paper is demonstrated.

Various multitasking approaches are investigated for optimization of large space structures. Judicious combination of microtasking, macrotasking, and autotasking is explored with the goal of achieving a vectorized and multitasked algorithm for optimization of a large structure with maximum speedup performance. Speedup results are presented and compared for three space truss structures with 526, 1046, and 3126 members.>

A practical solution to the structural analysis problem in a parallel processing environment is investigated through the use of the notion of cheap concurrency and the concept of thread. Portions of a structural analysis code implemented in C are parallelized using the encore parallel threads on an encore multimax multiprocessor computer. The issues of racing condition, synchronization, and mapping are considered and discussed. Results are reported on the effect of amount and frequency of shared memory access on the speedup. A discussion of the overhead time required for creating threads and a comparison of the overall computational time performance using two examples are presented.>