This paper proposes a generalized prediction system called a recurrent self-evolving fuzzy neural network (RSEFNN) that employs an on-line gradient descent learning rule to address the ...electroencephalography (EEG) regression problem in brain dynamics for driving fatigue. The cognitive states of drivers significantly affect driving safety; in particular, fatigue driving, or drowsy driving, endangers both the individual and the public. For this reason, the development of brain-computer interfaces (BCIs) that can identify drowsy driving states is a crucial and urgent topic of study. Many EEG-based BCIs have been developed as artificial auxiliary systems for use in various practical applications because of the benefits of measuring EEG signals. In the literature, the efficacy of EEG-based BCIs in recognition tasks has been limited by low resolutions. The system proposed in this paper represents the first attempt to use the recurrent fuzzy neural network (RFNN) architecture to increase adaptability in realistic EEG applications to overcome this bottleneck. This paper further analyzes brain dynamics in a simulated car driving task in a virtual-reality environment. The proposed RSEFNN model is evaluated using the generalized cross-subject approach, and the results indicate that the RSEFNN is superior to competing models regardless of the use of recurrent or nonrecurrent structures.
Hashing methods for efficient image retrieval aim at learning hash functions that map similar images to semantically correlated binary codes in the Hamming space with similarity well preserved. The ...traditional hashing methods usually represent image content by hand-crafted features. Deep hashing methods based on deep neural network (DNN) architectures can generate more effective image features and obtain better retrieval performance. However, the underlying data structure is hardly captured by existing DNN models. Moreover, the similarity (either visually or semantically) between pairwise images is ambiguous, even uncertain, to be measured in the existing deep hashing methods. In this article, we propose a novel hashing method termed deep fuzzy hashing network (DFHN) to overcome the shortcomings of existing deep hashing approaches. Our DFHN method combines the fuzzy logic technique and the DNN to learn more effective binary codes, which can leverage fuzzy rules to model the uncertainties underlying the data. Derived from fuzzy logic theory, the generalized hamming distance is devised in the convolutional layers and fully connected layers in our DFHN to model their outputs, which come from an efficient xor operation on given inputs and weights. Extensive experiments show that our DFHN method obtains competitive retrieval accuracy with highly efficient training speed compared with several state-of-the-art deep hashing approaches on two large-scale image datasets: CIFAR-10 and NUS-WIDE.
•The latest applications of deep learning in stock market prediction are presented.•The literature is reviewed with a general workflow for stock market prediction.•The often-ignored implementation ...and reproducibility in other surveys are examined.•The future directions along with the research frontiers are pointed out.
Stock market prediction has been a classical yet challenging problem, with the attention from both economists and computer scientists. With the purpose of building an effective prediction model, both linear and machine learning tools have been explored for the past couple of decades. Lately, deep learning models have been introduced as new frontiers for this topic and the rapid development is too fast to catch up. Hence, our motivation for this survey is to give a latest review of recent works on deep learning models for stock market prediction. We not only category the different data sources, various neural network structures, and common used evaluation metrics, but also the implementation and reproducibility. Our goal is to help the interested researchers to synchronize with the latest progress and also help them to easily reproduce the previous studies as baselines. Based on the summary, we also highlight some future research directions in this topic.
Deep artificial neural networks apply principles of the brain's information processing that led to breakthroughs in machine learning spanning many problem domains. Neuromorphic computing aims to take ...this a step further to chips more directly inspired by the form and function of biological neural circuits, so they can process new knowledge, adapt, behave, and learn in real time at low power levels. Despite several decades of research, until recently, very few published results have shown that today's neuromorphic chips can demonstrate quantitative computational value. This is now changing with the advent of Intel's Loihi, a neuromorphic research processor designed to support a broad range of spiking neural networks with sufficient scale, performance, and features to deliver competitive results compared to state-of-the-art contemporary computing architectures. This survey reviews results that are obtained to date with Loihi across the major algorithmic domains under study, including deep learning approaches and novel approaches that aim to more directly harness the key features of spike-based neuromorphic hardware. While conventional feedforward deep neural networks show modest if any benefit on Loihi, more brain-inspired networks using recurrence, precise spike-timing relationships, synaptic plasticity, stochasticity, and sparsity perform certain computation with orders of magnitude lower latency and energy compared to state-of-the-art conventional approaches. These compelling neuromorphic networks solve a diverse range of problems representative of brain-like computation, such as event-based data processing, adaptive control, constrained optimization, sparse feature regression, and graph search.
The trade-off between feature representation power and spatial localization accuracy is crucial for the dense classification/semantic segmentation of remote sensing images (RSIs). High-level features ...extracted from the late layers of a neural network are rich in semantic information, yet have blurred spatial details; low-level features extracted from the early layers of a network contain more pixel-level information but are isolated and noisy. It is therefore difficult to bridge the gap between high- and low-level features due to their difference in terms of physical information content and spatial distribution. In this article, we contribute to solve this problem by enhancing the feature representation in two ways. On the one hand, a patch attention module (PAM) is proposed to enhance the embedding of context information based on a patchwise calculation of local attention. On the other hand, an attention embedding module (AEM) is proposed to enrich the semantic information of low-level features by embedding local focus from high-level features. Both proposed modules are lightweight and can be applied to process the extracted features of convolutional neural networks (CNNs). Experiments show that, by integrating the proposed modules into a baseline fully convolutional network (FCN), the resulting local attention network (LANet) greatly improves the performance over the baseline and outperforms other attention-based methods on two RSI data sets.
The binary segmentation of roads in very high resolution (VHR) remote sensing images (RSIs) has always been a challenging task due to factors such as occlusions (caused by shadows, trees, buildings, ...etc.) and the intraclass variances of road surfaces. The wide use of convolutional neural networks (CNNs) has greatly improved the segmentation accuracy and made the task end-to-end trainable. However, there are still margins to improve in terms of the completeness and connectivity of the results. In this article, we consider the specific context of road extraction and present a direction-aware residual network (DiResNet) that includes three main contributions: 1) an asymmetric residual segmentation network with deconvolutional layers and a structural supervision to enhance the learning of road topology (DiResSeg); 2) a pixel-level supervision of local directions to enhance the embedding of linear features; and 3) a refinement network to optimize the segmentation results (DiResRef). Ablation studies on two benchmark data sets (the Massachusetts data set and the DeepGlobe data set) have confirmed the effectiveness of the presented designs. Comparative experiments with other approaches show that the proposed method has advantages in both overall accuracy and F1-score. The code is available at: https://github.com/ggsDing/DiResNet .
In recent years, neural network-based methods have shown promising results in hyperspectral image (HSI) denoising area. Real HSIs exhibit substantial variations in noise distribution due to various ...factors such as different imaging techniques, camera variations, imaging environments, and hardware aging. In this paper, we develop an eigenimage plus eigennoise level map guided convolutional neural network for HSI denoising. Our main idea is to perform eigendecomposition on HSIs, utilize the low-rank property of HSIs in the spectral dimension and approximate the spectral vectors in a low-dimensional orthogonal subspace, where representation coefficients are called eigenimages. Besides eigenimages, we make use of estimated eigennoise level map as an input to guide the network for denoising. The proposed network can be constructed without restriction in the number of eigencomponents by using all eigenimages and eigennoise level maps of training noisy-clean pairs. In the inference part, the trained network can be used to remove noise in observed eigenimages without restriction in the number of eigencomponents, and an underlying clean image HSI can be estimated by performing orthogonal projection back. Experimental results on both simulated and real HSIs demonstrate the effectiveness of our trained Eigen-CNN compared with state-of-the-art HSI denoising methods. A MATLAB demo of this work is available at https://github.com/LinaZhuang/HSI-denoiser-Eigen-CNN for the sake of reproducibility.
Programmable metasurface (PM) exhibits powerful capabilities to manipulate electromagnetic (EM) waves with controlled active components loaded on the sub-wavelength elements. However, for ...electrically large metasurface with complex structures, it is difficult to implement the fast mapping from the codes, i.e. states of the active components, to radiation patterns and vice versa. In this paper, artificial neural networks are employed to realize code-to-pattern and pattern-to-code mapping accurately and efficiently. As for the code-to-pattern (C-P) mapping, a novel physics-inspired neural network (PINN), which is primarily inspired by the discrete dipole approximation (DDA) method, is proposed. The PINN is physically interpretable and shows a strong few-shot learning ability. The average error between the patterns predicted by the PINN and measured patterns is as low as 2.31 dB. For the pattern-to-code (P-C) mapping, a deep neural network (DNN) is proposed with PINN as its teacher network, i.e. PINN is used to generate more radiation patterns to train the DNN. The average accuracy of codes prediction is higher than 98.4%. Finally, an intelligent beamforming scheme is implemented by combining the PINN and DNN. For the desired patterns, the required codes could be accurately calculated by the DNN in real time. Then the synthesized patterns of the required codes could be achieved by the PINN to compare with the desired patterns. The proposed scheme is a first step toward practical applications of PM in fields of sensing and communication.
Deep learning (DL) is a new machine learning (ML) methodology that has found successful implementations in many application domains. However, its usage in communications systems has not been well ...explored. This paper investigates the use of the DL in modulation classification, which is a major task in many communications systems. The DL relies on a massive amount of data and, for research and applications, this can be easily available in communications systems. Furthermore, unlike the ML, the DL has the advantage of not requiring manual feature selections, which significantly reduces the task complexity in modulation classification. In this paper, we use two convolutional neural network (CNN)-based DL models, AlexNet and GoogLeNet. Specifically, we develop several methods to represent modulated signals in data formats with gridlike topologies for the CNN. The impacts of representation on classification performance are also analyzed. In addition, comparisons with traditional cumulant and ML-based algorithms are presented. Experimental results demonstrate the significant performance advantage and application feasibility of the DL-based approach for modulation classification.
Fault diagnosis, which identifies the root cause of the observed out-of-control status, is essential to counteracting or eliminating faults in industrial processes. Many conventional data-driven ...fault diagnosis methods ignore the fault tendency of abnormal samples, and they need a complete retraining process to include the newly collected abnormal samples or fault classes. In this article, a broad convolutional neural network (BCNN) is designed with incremental learning capability for solving the aforementioned issues. The proposed method combines several consecutive samples as a data matrix, and it then extracts both fault tendency and nonlinear structure from the obtained data matrix by using convolutional operation. After that, the weights in fully connected layers can be trained based on the obtained features and their corresponding fault labels. Because of the architecture of this network, the diagnosis performance of the BCNN model can be improved by adding newly generated additional features. Finally, the incremental learning capability of the proposed method is also designed, so that the BCNN model can update itself to include new coming abnormal samples and fault classes. The proposed method is applied both to a simulated process and a real industrial process. Experimental results illustrate that it can better capture the characteristics of the fault process, and effectively update diagnosis model to include new coming abnormal samples, and fault classes.