Combining physics and deep learning to learn continuous-time dynamics models

Deep learning (also known as deep structured learning, hierarchical learning, or deep machine learning) is a branch of machine learning based on a set of algorithms that attempt to model high-level abstractions in data by using a deep graph with multiple processing layers, composed of multiple linear and nonlinear transformations. Deep learning has been characterized as a class of machine learning algorithms with the following characteristics [257]: • They use a cascade of many layers of nonlinear processing units for feature extraction and transformation. Each successive layer uses the output from the previous layer as input. The algorithms may be supervised or unsupervised and applications include pattern analysis (unsupervised) and classification (supervised). • They are based on the (unsupervised) learning of multiple levels of features or representations of the data. Higher-level features are derived from lower-level features to form a hierarchical representation. • They are part of the broader machine learning field of learning representations of data. • They learn multiple levels of representations that correspond to different levels of abstraction; the levels form a hierarchy of concepts. These definitions have in common: multiple layers of nonlinear processing units and the supervised or unsupervised learning of feature representations in each layer, with the layers forming a hierarchy from low-level to high-level features. Various deep learning architectures such as deep neural networks, convolutional deep neural networks, deep belief networks (DBN), and recurrent neural networks have been applied to fields like computer vision, automatic speech recognition, natural language processing, audio recognition, and bioinformatics where they have been shown to produce state-of-the-art results on various tasks. In this chapter we start in Section 7.1 with an introduction, giving a brief history of this field, the relevant literature, and its applications. Then we study some basic concepts of deep learning such as convolutional neural networks, recurrent neural networks, backpropagation algorithm, restricted Boltzmann machines, and deep learning networks in Section 7.2. Then we illustrate three examples for Apache Spark implementation for mobile big data (MBD), user moving pattern extraction, and combination with nonparametric Bayesian learning, respectively, in Sections 7.3 through 7.5. Finally, we have summary in Section 7.6.

Demosaicking is reconstructing a RGB image from the mosaicked image recorded by a single sensor with a color filter array (CFA), which has achieved great progress via deep learning. Some works on joint CFA design and demosaicking via deep learning have been presented. However, almost all existing approaches focus only on optimizing filter arrangements without considering the used spectral sensitivity functions (SSFs). The commonly used SSFs are to mimic human trichromatic perception, which are not optimal for deep-learning-based de-mosaicking. In this paper, we simultaneously learn the CFA and demosaicking using a deep convolutional neural network, where both the filter arrangement and used SSFs are optimized. By modeling the applying of SSFs as a convolutional layer with physical constraints, we formulate our joint learning approach as an end-to-end autoencoder, which can be trained as the standard convolutional neural networks. Our approach allows the designing of CFA patterns with arbitrary sizes, where the filter arrangements can be predefined or not. We demonstrate the effectiveness and advantages of our approach by comparing with fixed Gaussian functions and various predefined filter arrangements.

Deep learning is a recently developed feature representation technique for data with complicated structures, which has great potential for soft sensing of industrial processes. However, most deep networks mainly focus on hierarchical feature learning for the raw observed input data. For soft sensor applications, it is important to reduce irrelevant information and extract quality-relevant features from the raw input data for quality prediction. To deal with this problem, a novel deep learning network is proposed for quality-relevant feature representation in this article, which is based on stacked quality-driven autoencoder (SQAE). First, a quality-driven autoencoder (QAE) is designed by exploiting the quality data to guide feature extraction with the constraint that the potential features should largely reconstruct the input layer data and the quality data at the output layer. In this way, quality-relevant features can be captured by QAE. Then, by stacking multiple QAEs to construct the deep SQAE network, SQAE can gradually reduce irrelevant features and learn hierarchical quality-relevant features. Finally, the high-level quality-relevant features can be directly applied for soft sensing of the quality variables. The effectiveness and flexibility of the proposed deep learning model are validated on an industrial debutanizer column process.

Short-term power load prediction is not only a critical part of power system dispatching but also an essential task for power marketing, grid planning, and other management departments; deep learning is an artificial intelligence method that has gained extensive attention in recent years; this paper selected three models with the most typical signs of deep learning recurrent neural networks and studied their model performance for short-term power load predictions, including power load data preprocessing, feature selection of prediction models, determination of model parameters, and comparison of prediction characteristics between shallow and deep networks, to explore and prove the applicability of deep learning neural networks in short-term power load predictions, ultimately improving the accuracy of short-term load predictions.KeywordsPower loadDeep learningPrediction

Deep neural networks have made significant achievements in representation learning of traditionally man-made features, especially in terms of complex objects. Over the decades, this learning process has attracted thousands of researchers and has been widely used in the speech, visual, and text recognition fields. One deep network multi-layer extreme learning machine (ML-ELM) achieves a good performance in representation learning while inheriting the advantages of faster learning and the approximating capability of the extreme learning machine (ELM). However, as with most deep networks, the ML-ELM’s algorithmic performance largely depends on the probability distribution of the training data. In this paper, we propose an improved ML-ELM made via using the local significant regions at the input end to enhance the contributions of these regions according to the idea of the selective attention mechanism. To avoid involving and exploring the complex principle of the attention system and to focus on the clarification of our local regional enhancement idea, the paper only selects two typical attention regions. One is the geometric central region, which is normally the important region to attract human attention due to the focal attention mechanism. The other is the task-driven interest region, with facial recognition as an example. The comprehensive experiments are done on the three public datasets of MNIST, NORB, and ORL. The comparison experiment results demonstrate that our proposed region-enhanced ML-ELM (RE-ML-ELM) achieves performance increases in important feature learning by utilizing the apriori knowledge of attention and has a higher recognition rate than that of the normal ML-ELM and the basic ELM. Moreover, it benefits from the non-iterative parameter training method of other ELMs, and our proposed algorithm outperforms most state-of-the-art deep networks such as deep belief network(DBN), in the aspects of training efficiency. Furthermore, because of the deep structure with fewer hidden nodes at each layer, our proposed RE-ML-ELM achieves a comparable training efficiency to that of the ML-ELM but has a higher training speed with the basic ELM, which is normally the width single network that has more hidden nodes to obtain the similar recognition accuracy with the deep networks. Based on our idea of combining the apriori knowledge of the human selective attention system with the data learning, our proposed region-enhanced ML-ELM increases the image classification performance. We believe that the idea of intentionally combining psychological knowledge with the most algorithms based on data-driven learning has the potential to improve their cognitive computing ability.

In recent decades, deep learning (DL) has become a rapidly growing research direction, redefining the state-of-the-art performances in a wide range of techniques, such as object detection and speech recognition. In the aircraft design, dynamics, and control field, many works hinge on the information-rich data-driven approach, which includes the fusion-based prognostic and health management, the airliner's flight safety monitoring, intelligent sensing, and flight control systems development. While DL provides great potentials to solve these data-driven problems, a systematic review and discussion as to how the DL has been/can be used for these problems are still missing in relation to the rapidly developing and widely used DL techniques. In this article, we aim to address this urgent issue to provide a timely overview of the state-of-the-art for applying DL to the aircraft design, dynamics, and control field. In particular, we briefly introduce five representative DL methods, i.e., deep neural network, deep autoencoder, deep belief network, convolutional neural network, and recurrent neural network. Mathematical definitions for each method are presented, and illustrative applications are also discussed. We then review the existing DL-based works that have appeared in the aircraft design, dynamics, and control field. The review efforts are divided into two major groups, i.e., the own-ship aircraft modeling, wherein the works have been/can be implemented online for the aircraft design/dynamics/control, and other airplanes research works, wherein DL-based schemes provide offline monitoring of the aircraft operation. We then summarize the data sources and DL architectures. Referring to the experiences of DL research works/techniques development in other related fields, future opportunities, challenges, and potential solutions for implementing DL in the aircraft design, dynamics, and control field are also discussed.

Deep learning has been applied to saliency detection in recent years. The superior performance has proved that deep networks can model the semantic properties of salient objects. Yet it is difficult for a deep network to discriminate pixels belonging to similar receptive fields around the object boundaries, thus deep networks may output maps with blurred saliency and inaccurate boundaries. To tackle such an issue, in this work, we propose a deep Level Set network to produce compact and uniform saliency maps. Our method drives the network to learn a Level Set function for salient objects so it can output more accurate boundaries and compact saliency. Besides, to propagate saliency information among pixels and recover full resolution saliency map, we extend a superpixel-based guided filter to be a layer in the network. The proposed network has a simple structure and is trained end-to-end. During testing, the network can produce saliency maps by efficiently feedforwarding testing images at a speed over 12FPS on GPUs. Evaluations on benchmark datasets show that the proposed method achieves state-of-the-art performance.

Recently, data-driven algorithms such as deep neural networks have attracted a lot of attention and have become a popular area of research and development. This interest is driven by several factors, such as recent advances in processing power (cheap and powerful hardware), the availability of large datasets (big data), and several small but important algorithmic advances (e.g., convolutional layers). Nowadays, deep neural networks are the state-of-the-art for several computer vision tasks, such as the ones that require high-level understanding of image semantics, e.g., image classification, object segmentation, saliency detection, and also in low level image processing tasks, such as image denoising, inpainting, and super-resolution, among others. Deep learning can handle large volumes of video data by making use of its powerful non-linear mapping and extracting high-level features with very deep networks. Incorporating deep learning into the IoVT can provide radical innovations in video sensing, coding, enhancing, understanding, and evaluation areas making it easier to handle the enormous growth in video data compared to traditional methods.

Hyperspectral remote sensing image classification based on residual generative Adversarial Neural Networks

Deep (machine) learning in recent years has significantly increased the predictive modeling strengths for applications in the areas of computer vision, speech recognition, and natural language processing. Being heavily data-driven, conventional general-purpose deep learning methods mainly rely on training data to learn task-specific data representations. In this thesis, we focus on leveraging novel learning algorithms in deep learning to tackle visual recognition challenges at the three key levels -- namely pixel, object, and image levels. Throughout the thesis, we demonstrate how each and every of our proposed methods can address a specific visual recognition problem. Firstly, we apply deep learning to improve the saliency detection performance of deep networks. A coarse saliency map of the input image's salient object is initially predicted by a convolutional-deconvolutional network (or equivalently a fully convolutional network). Following that, the coarse saliency map is sequentially refined by a separate network called Recurrent Attentional Convolutional-Deconvolutional Network (RACDNN) that shares its network weights across its sequential steps. At every sequential (temporal) step, the same RACDNN is applied to a newly selected/attended image sub-region and it performs saliency map refinement for that particular sub-region. Despite having many learnable network parameters and no access to additional training data, RADCNN can be trained well and it outperforms the "single-step" (coarse saliency detection) baseline and state-of-the-art saliency detection methods by large margins, on several saliency detection benchmark datasets. Secondly, we introduce auto-encoder weight transfer networks (AE-WTN+), a deep model for scaling up object detection. Conventional WTN transfers object class knowledge from the classification weights of a pretrained large-scale image classification network, to the classification weights of an object detection network. However, they suffer from under-fitting and are not able to preserve the rich object class information of novel classes. In this work, we propose a new WTN method -- AE-WTN+ which makes use of normalization techniques to ease network optimization. Inputs and intermediate features are normalized to improve trainability for reduced under-fitting. More importantly, a novel auto-encoding (reconstruction) loss is proposed to encourage information preservation of all classes in the weight predictions of AE-WTN+. This enables the predicted weights of AE-WTN+ to maintain the class discriminability and neighbourhood relationships contained in the pre-trained large-scale classification weights. Experiments on large-scale detection datasets validate the effectiveness of AE-WTN+ on both seen and novel classes. Thirdly, we improve the efficiency of cross-layer network connections for very deep convolutional network in the context of image classification. This gives rise to a new convolutional network architecture called Deluge Networks (DelugeNets) which pass the output (feature activations) of any preceding composite network layer as direct inputs to all succeeding composite layers. For the incoming feature activations of preceding composite layers, the succeeding composite layers first apply rather lightweight cross-layer depthwise convolutional weights on them for feature aggregation. The heavy lifting is carried out by the last convolutional weight layer in each preceding composite layer that performs cross-channel convolutions. The proposed technique enables unimpeded information flow among many layers in very deep DelugeNets with great efficiency. Experiments on small-scale and large-scale datasets show that DelugeNets perform competitively to state-of-the-art deep convolutional networks at lower computational and parameter costs. Fourthly, we propose Stochastic Downsampling (SDPoint), a method to overcome a major limitation suffered by conventional deep convolutional networks -- their computational costs during inference are fixed and cannot be changed. SDPoint downsamples the feature maps at a randomly selected point (layer index) in the network hierarchy during training. The many downsampling configurations known as SDPoint instances each entail a unique downsampling point and downsampling ratio. The same convolutional network weights are shared across the different SDPoint instances that downsample feature maps differently. During inference, the computational cost of a network trained with SDPoint can be instantaneously adjusted to fit a given computational budget, by manually selecting an appropriate SDPoint instance. Additionally, sharing weights between different feature map scales provides significant regularization benefits, making the networks trained with SDPoint less sensitive to scales and more competent. In image classification experiments, SDPoint demonstrates significant cost-accuracy and computational advantages, over independently-trained (non-weight-shared) baselines with various inference costs.

Purpose: The aim of the article is to develop an algorithm for classifying cracks in the analyzed images using modern methods of deep machine learning and transfer learning based on pretrained convolutional neural network - Inception-ResNet-v2. Design/Methodology/Approach: Transfer learning based on the pretrained convolutional neural network was used to categorize the images. The fully conected layer of the Inception-ResNet-v2 network has been modified. The last layer was trained using a two-class (binary) linear SVM (Support Vector Machine). In total, 20,000 training cases (images) were used to train the fully connected layer within transfer learning process. The research analyzed the possibility of using the deep neural networks for quick and fully automatic identification of cracks / defects on the surface of analyzed parts. Findings: The results indicate that pretrained convolutional neural network using SVM to train a fully connected layer is a very effective solution for visual crack / fault detection. In the analyzed model, a positive classification was obtained at the level of 99.89%. Practical Implications: The model presented in the article can be used in quality control carried out by process monitoring. An effective model for identifying defective parts can be used in both logistics and production processes. Originality/Value: A novelty is the use of a freely available, deep neural network trained to classify 1000 categories of various images for binary categorization of faults (cracks). The algorithm was adjusted by replacing the primary, 1000-output fully connected layer in the Inception-ResNet-v2 network with a binary layer (2 categories). The fully connected layer has been trained using the classification version of the popular SVM learner, but thanks to the combination of this layer with the sophisticated fearure extraction ability of the pre-trained Inception-ResNet-v2 deep network, the resulting predictive model enables the classification of defects with a very high level of accuracy.

Deep learning neural network approach to thermal-wave imaging of damage in solids with application to diffusivity measurements of a green (unsintered) metal powder compact slab

DBMiP: A pre-training method for information propagation over deep networks

In the power system operational/planning studies, it is a crucial task to provide the load consumption information in the look‐ahead times. The huge variation of the power system infrastructure in recent years has led to significant changes in the consumers’ consumption pattern. Therefore, short‐term load forecasting (STLF) is transformed to a more complicated problem in recent years. To address this issue, this paper proposes a graph‐based deep neural network to capture full spatial‐temporal features and be able to oversee high volatility time series including load sequence. The proposed spatial deep learning structure benefits from learning the spatial feature using Gabor filter‐oriented layers and full understanding the temporal behaviour based on bidirectional networks. The designed learning‐based system is developed as a graph‐based learning system to improve the accuracy considering the meteorological information behaviour. To verify the performance of the designed deep graph network, the actual load data of Shiraz, Iran, is used. Besides, to demonstrate the superiority and effectiveness of the proposed, the designed deep graph network is compared with three well‐known shallow and deep networks in different cases including yearly performance, seasonal performance, and sensitivity analysis on the meteorological data.

Deep learning has been proved effective in multiple object tracking, which confronts the difficulties of frequent occlusions, confusing appearance, in‐and‐out objects, and lack of enough labelled data. Recently, deep learning based multi‐object tracking methods make a rapid progress from representation learning to network modelling due to the development of deep learning theory and benchmark setup. In this study, the authors summarise and analyse deep learning based multi‐object tracking methods which are top‐ranked in the public benchmark test. First, they investigate functionality of deep networks in these methods, and classify the methods into three categories as description enhancement using deep features, deep network embedding, and end‐to‐end deep network construction. Second, they review deep network structures in these methods, and detail the usage and training of these networks for multi‐object tracking problem. Through experimental comparison of tracking results in the benchmarks in total and by group, they finally show the effectiveness of deep networks for tracking employed in different manners, and compare the advantages of these networks and their robustness under different tracking conditions. Moreover, they analyse the limitations of current methods, and draw some useful conclusions to facilitate the exploration of new directions for multi‐object tracking.