Search

Summary: Lithium‐ion battery state‐of‐health (SOH) estimation and remaining usable life (RUL) prediction are important for battery prognosis and health management. In this article, a framework‐combined deep convolution neural network (DCNN) with double‐layer long short‐term memory (LSTM) is proposed, which is designed for online health prognosis. Based on the raw data obtained during the constant current charging process, the aging characteristics of the battery can be extracted by DCNN to estimate SOH. The estimation results are then sent to the LSTM for the temporal prediction of the RUL. This framework considers both temporal and spatial characteristics of data, and the powerful spatial feature extraction ability of DCNN and the effectiveness of LSTM for time series problems can ensure the high precision of calculation. At the same time, the hyperparameters of the neural networks, which can highly affect the performance of networks, are obtained by Bayesian optimization to ensure the networks run in the best status. The results show that the proposed method can achieve low root mean square error, which are inferior to 0.0061 and 0.0627 for SOH estimation and RUL prediction, respectively. [ABSTRACT FROM AUTHOR]

Video captioning is a challenging task that aims to generate linguistic description based on video content. Most methods only incorporate visual features (2D/3D) as input for generating visual and non-visual words in the caption. However, generating non-visual words usually depends more on sentence-context than visual features. The wrong non-visual words can reduce the sentence fluency and even change the meaning of sentence. In this paper, we propose a syntax-guided hierarchical attention network (SHAN), which leverages semantic and syntax cues to integrate visual and sentence-context features for captioning. First, a globally-dependent context encoder is designed to extract the global sentence-context feature that facilitates generating non-visual words. Then, we introduce hierarchical content attention and syntax attention to adaptively integrate features in terms of temporality and feature characteristics respectively. Content attention helps focus on time intervals related to the semantic of current word, while cross-modal syntax attention uses syntax information to model importance of different features for target word’s generation. Moreover, such hierarchical attention can enhance the model interpretability for captioning. Experiments on MSVD and MSR-VTT datasets show the comparable performance of our method compared with current methods. [ABSTRACT FROM AUTHOR]

As a deep learning model, a deep belief network (DBN) consists of multiple restricted Boltzmann machines (RBMs). Based on DBN, many intelligent fault diagnosis methods are proposed. However, these methods seldom considered the appearance of new working conditions during the operation of real machines. Varying working conditions lead to a change of feature distributions and finally result in low diagnosis accuracies. Therefore, we propose a distribution-invariant DBN (DIDBN) to learn distribution-invariant features directly from raw vibration data. DIDBN consists of a locally connected RBM (LCRBM) layer, a fully connected RBM layer, and an RBM layer with a mean discrepancy maximum (MDM-RBM). The LCRBM layer is designed with a locally connected structure. By proposing MDM, the MDM-RBM layer is able to obtain features that have close distributions under varying working conditions. Followed by a softmax classifier, DIDBN is able to recognize faults. The proposed method is applied to two diagnosis cases. Results verify that DIDBN is able to learn distribution-invariant features and achieve higher diagnosis accuracies than recently proposed methods. Moreover, the reason why DIDBN is able to learn distribution-invariant features is explained by visualizing the feature learning process. [ABSTRACT FROM AUTHOR]

Cross-modal retrieval has attracted intensive attention in recent years, where a substantial yet challenging problem is how to measure the similarity between heterogeneous data modalities. Despite using modality-specific representation learning techniques, most existing shallow or deep models treat different modalities equally and neglect the intrinsic modality heterogeneity and information imbalance among images and texts. In this paper, we propose an online similarity function learning framework to learn the metric that can well reflect the cross-modal semantic relation. Considering that multiple CNN feature layers naturally represent visual information from low-level visual patterns to high-level semantic abstraction, we propose a new asymmetric image–text similarity formulation which aggregates the layer-wise visual–textual similarities parameterized by different bilinear parameter matrices. To effectively learn the aggregated similarity function, we develop three different similarity combination strategies, i.e., average kernel, multiple kernel learning, and layer gating. The former two kernel-based strategies assign uniform weights on different layers to all data pairs; the latter works on the original feature representation and assigns instance-aware weights on different layers to different data pairs, and they are all learned by preserving the bi-directional relative similarity expressed by a large number of cross-modal training triplets. The experiments conducted on three public datasets well demonstrate the effectiveness of our methods. [ABSTRACT FROM AUTHOR]

Battery fault diagnosis is essential to ensure the safe operation of electric vehicles (EVs). In this paper, due to the complexity of EVs' battery thermal runaway tracing investigation and the limited capacity of on-board computing system, a double-layer fault diagnosis strategy for abnormal cells is proposed. The method bases on probability distribution, which can accurately trace a faulty cell and avoid misinterpreting a normal cell. In this method, unified statistical features are extracted from the big data during vehicle charging, and the corresponding statistical values are analyzed based on Gaussian mixture model and abnormal alarm is made based on the risk accumulation in double-layer diagnostics. The electric vehicles with thermal runaway accident are taken as examples to verify the method, and based on the data of normal-running vehicles, the false alarm tests are carried out. The verification results show that the proposed method can not only successfully identify the outlier cells, but also not generate false alarm, which is conducive to the practical application of fault diagnosis in the on-board battery management system. • Gaussian mixture model is used for battery fault diagnosis. • The unified statistical values are extracted to reduce the requirement of data. • Thresholds by using risk accumulation is proposed to detect the abnormal cells. • The false alarm test is performed as an auxiliary verification for the algorithm. [ABSTRACT FROM AUTHOR]

Vibration signals of faulty rolling element bearings usually exhibit non-linear and non-stationary characteristics caused by the complex working environment. It is difficult to develop a robust method to detect faults in bearings based on signal processing techniques. In this paper, convolutional neural network -based hidden Markov models (CNN HMMs) are presented to classify multi-faults in mechanical systems. In CNN HMMs, a CNN model is first employed to learn data features automatically from raw vibration signals. By utilizing the t-distributed stochastic neighbor embedding (t-SNE) technique, feature visualization is constructed to manifest the powerful learning ability of CNN. Then, HMMs are employed as a strong stability tool to classify faults. Both the benchmark data and experimental data are applied to the CNN HMMs. Classification results confirm the superior performance of the present combination model by comparing with CNN model alone, support vector machine (SVM) and back propagation (BP) neural network. It is shown that the average classification accuracy ratios are 98.125% and 98% for two data series with agreeable error rate reductions. [ABSTRACT FROM AUTHOR]

Different kinds of features describe different aspects of image data, and each feature can be treated as a view when we take it as a particular understanding of images. Leveraging multiple views provides a richer and comprehensive description than using only a single view. However, multiview data are often represented by high-dimensional heterogeneous features, so it is meaningful to find a low-dimensional consensus representation from multiple views. In this paper, we propose an unsupervised multiview dimensionality reduction method for images based on bilevel latent space learning. As different views have different physical meanings and statistical properties, they are not directly comparable. Therefore, we learn the comparable representation for each view in the first level. The shared and the private nature of multiview data are exploited to accurately preserve the information of each view. Then, we fuse different views into a low-dimensional representation by conducting joint matrix factorization in the second level. To guarantee the low-dimensional representation to be compact and discriminative, the intrinsic geometric structure of data is utilized. Besides, our method considers resisting the outliers and noise contained in multiview data, which may influence the learned representation and deteriorate its semantic consistency. We design appropriate optimization objectives to learn the latent spaces in different levels. Compared with the existing methods, our method could provide a more flexible multiview learning strategy that not only accurately captures the information of each view but also is robust to outliers and noise, which can obtain a more discriminative and compact low-dimensional representation. Experiments on two real-world image data sets demonstrate the advantages of our method over the existing multiview dimensionality reduction methods. [ABSTRACT FROM AUTHOR]

In this paper, we detect communities from trajectories. Existing algorithms for trajectory clustering usually rely on simplex representation and a single proximity-related metric. Unfortunately, additional information markers (e.g., social interactions or semantics in the spatial layout) are ignored, leading to the inability to fully discover the communities in trajectory database. This is especially true for human-generated trajectories, where additional fine-grained markers (e.g., movement velocity at certain locations, or the sequence of semantic spaces visited) are especially useful in capturing latent relationships among community members. To overcome this limitation, we propose TODMIS, a general framework for Trajectory-based cOmmunity Detection by diffusion modeling on Multiple Information Sources. TODMIS combines additional information with raw trajectory data and construct the diffusion process on multiple similarity metrics. It also learns the consistent graph Laplacians by constructing the multi-modal diffusion process and optimizing the heat kernel coupling on each pair of similarity matrices from multiple information sources. Then, dense sub-graph detection is used to discover the set of distinct communities (including community size) on the coupled multi-graph representation. At last, based on the community information, we propose a novel model for online recommendation. We evaluate TODMIS and our online recommendation methods using different real-life datasets. Experimental results demonstrate the effectiveness and efficiency of our methods. [ABSTRACT FROM AUTHOR]

Due to the prevalence of We-Media, information is quickly published and received in various forms anywhere and anytime through the Internet. The rich cross-media information carried by the multimodal data in multiple media has a wide audience, deeply reflects the social realities, and brings about much greater social impact than any single media information. Therefore, automatically detecting topics from cross media is of great benefit for the organizations (i.e., advertising agencies and governments) that care about the social opinions. However, cross-media topic detection is challenging from the following aspects: 1) the multimodal data from different media often involve distinct characteristics and 2) topics are presented in an arbitrary manner among the noisy web data. In this paper, we propose a multimodality fusion framework and a topic recovery (TR) approach to effectively detect topics from cross-media data. The multimodality fusion framework flexibly incorporates the heterogeneous multimodal data into a multimodality graph, which takes full advantage from the rich cross-media information to effectively detect topic candidates (T.C.). The TR approach solidly improves the entirety and purity of detected topics by: 1) merging the T.C. that are highly relevant themes of the same real topic and 2) filtering out the less-relevant noise data in the merged T.C. Extensive experiments on both single-media and cross-media data sets demonstrate the promising flexibility and effectiveness of our method in detecting topics from cross media. [ABSTRACT FROM AUTHOR]

Abstract: Currently, Nearest-Neighbor approaches (NN) have been applied to large scale real world image data mining. However, the following three disadvantages prevent them from wider application compared to other machine learning methods: (i) the performance is inferior on small datasets; (ii) the performance will degrade for data with high dimensions; (iii) they are heavily dependent on the chosen feature and distance measure. In this paper, we try to overcome the three mentioned intrinsic weaknesses by taking the abundant and diversified content of social media images into account. Firstly, we propose a novel neighborhood similarity measure which encodes both the local density information and semantic information, thus it has better generalization power than the original image-to-image similarity. Secondly, to enhance the scalability, we adopt kernelized Locality Sensitive Hashing (KLSH) to conduct approximated nearest neighbor search by utilizing a set of kernels calculated on several complementary image features. Finally, to enhance the robustness on diversified genres of images, we propose to fuse the discrimination power of different features by combining multiple neighborhood similarities calculated on different features/kernels with the entire retrieved nearest labeled and unlabeled image via the hashing systems. Experimental results on visual categorization on the Caltech-256 and two social media databases show the advantage of our method over traditional NN methods using the labeled data only. [Copyright &y& Elsevier]

This paper proposes a novel method named Multi-label Double-layer Learning (MDLL) for multi-label cross-modal retrieval task. MDLL includes two stages (layers): L2C (Label to Common) and C2L (Common to Label). In the L2C stage, considering that labels can provide semantic information, we take label information as an auxiliary modality and apply a covariance matrix to represent label similarity in multi-label situation. Thus we can maximize the correlation of different modalities and reduce their semantic gap in the L2C stage. In addition, we find that samples with the same semantic labels may have different contents from users’ view. According to this problem, in the C2L stage, labels are projected to a latent space learned from features of image and text. By this way, the label latent space are more related to the sample’s contents. Then, it is noticed that the samples have same labels but various contents can be decreased. In MDLL, iterative learning of the L2C and C2L stages will improve the discriminative ability greatly and decline the discrepancy between the labels and the contents. To show the effectiveness of MDLL, some experiments are conducted on three multi-label cross-modal retrieval tasks (Pascal Voc 2007, Nus-wide, and LabelMe), on which competitive results are obtained. [ABSTRACT FROM AUTHOR]