With the development of the data-driven modeling techniques, using the neural network to simulate the transport process of atmospheric pollutants and constructing PM2.5 time-series prediction model have become a hot t...
详细信息
With the development of the data-driven modeling techniques, using the neural network to simulate the transport process of atmospheric pollutants and constructing PM2.5 time-series prediction model have become a hot topic. The existing data-driven approaches often ignore the dynamical relationships among multiple sites in urban areas, which results in nonideal prediction accuracy. In response to this problem, this article proposes a long short-term memory (LSTM) autoencoder multitask learning model to predict PM2.5 time series in multiple locations city wide. The model could implicitly and automatically excavate the intrinsic relevance among the pollutants in different stations. And the meteorological information from the monitoring stations is fully utilized, which is beneficial for the performance of the proposed model. Specifically, multilayer LSTM networks can simulate the spatiotemporal characteristics of urban air pollution particles. And using the stacked autoencoder to encode the key evolution pattern of urban meteorological systems could provide important auxiliary information for PM2.5 time-series prediction. In addition, multitask learning could automatically discover the dynamical relationship between multiple key pollution time series and solve the problem of insufficient use of multisite information in the modeling process of the traditional data-driven methods. The simulation results of PM2.5 prediction in Beijing indicate the effectiveness of the proposed method.
AUTOVC is a voice-conversion method that performs self-reconstruction using an autoencoder structure for zero-shot voice conversion. AUTOVC has the advantage of being easy and simple to learn because it only uses the ...
详细信息
AUTOVC is a voice-conversion method that performs self-reconstruction using an autoencoder structure for zero-shot voice conversion. AUTOVC has the advantage of being easy and simple to learn because it only uses the autoencoder loss for learning. However, it performs voice conversion by disentangling speech information from speakers and linguistic information by adjusting the bottleneck dimension;this requires highly meticulous fine tuning of the bottleneck dimension and involves a tradeoff between speech quality and speaker similarity. To address these issues, neural analysis and synthesis (NANSY)-a fully self-supervised learning system that uses perturbations to extract speech features-is proposed. NANSY solves the problem of the adjustment of the bottleneck dimension by utilizing perturbation and exhibits high-reconstruction performance. In this study, we propose perturbation AUTOVC, a voice conversion method that utilizes the structure of AUTOVC and the perturbation of NANSY. The proposed method applies perturbations to speech signals (such as NANSY signals) to solve the problem of the voice conversion method using bottleneck dimensions. Perturbation is applied to remove the speaker-dependent information present in the speech, leaving only the linguistic information, which is then passed through a content encoder and modeled as a content embedding containing only the linguistic information. To obtain speaker information, we used x-vectors, which are extensively used in pretrained speaker recognition. The concatenated linguistic and speaker information extracted from the encoder and additional energy information is used as input to the decoder to perform self-reconstruction. Similar to AUTOVC, it is easy and simple to learn using only the autoencoder loss. For the evaluation, we measured three objective evaluation metrics: character error rate (%), cosine similarity, and short-time objective intelligibility, as well as a subjective evaluation metric: mean opinio
With the proliferation of the Internet of Things, a large amount of multivariate time series (MTS) data is being produced daily by industrial systems, corresponding in many cases to life-critical tasks. The recent ano...
详细信息
With the proliferation of the Internet of Things, a large amount of multivariate time series (MTS) data is being produced daily by industrial systems, corresponding in many cases to life-critical tasks. The recent anomaly detection researches focus on using deep learning methods to construct a normal profile for MTS. However, without proper constraints, these methods cannot capture the dependencies and dynamics of MTS and thus fail to model the normal pattern, resulting in unsatisfactory performance. This paper proposes CAE-AD, a novel contrastive autoencoder for anomaly detection in MTS, by introducing multi -grained contrasting methods to extract normal data pattern. First, to capture the temporal dependency of series, a projection layer is employed and a novel contextual contrasting method is applied to learn the robust temporal representation. Second, the projected series is transformed into two different views by using time-domain and frequency-domain data augmentation. Last, an instance contrasting method is proposed to learn local invariant characteristics. The experimental results show that CAE-AD achieves an F1-score ranging from 0.9119 to 0.9376 on the three public datasets, outperforming the baseline methods.(c) 2022 Published by Elsevier Inc.
The spatial heterogeneity of geochemical background is often ignored in geochemical anomaly recognition, leading to ineffective recognition of valuable anomalies for geochemical prospecting. In this paper, a Spatially...
详细信息
The spatial heterogeneity of geochemical background is often ignored in geochemical anomaly recognition, leading to ineffective recognition of valuable anomalies for geochemical prospecting. In this paper, a Spatially Constrained Multi-autoencoder (SCMA) approach is proposed to deal with such an issue in multivariate geochemical anomaly recognition, which includes two unique steps: (1) with the consideration of both chemical similarity and spatial continuity of geochemical samples, a region is divided into multiple sub-domains to discriminate the various backgrounds over space, through multivariate clustering, spatial filtering, and spatial fusion;and (2) the geochemical background of each sub-domain is learned and reconstructed by a multi-autoencoder structure, which is designed to reduce the effects of random initialization of weights in an autoencoder neural network. Finally, the anomaly score is calculated as the difference between the observed geochemical features and the reconstructed features. The performance of SCMA was demonstrated by a case study involving Cu, Mn, Pb, Zn and Fe2O3 in stream sediment samples from the Chinese National Geochemical Mapping Project, in the southwestern Fujian province of China. The results showed that the spatial domain constraining greatly improved the quality of anomaly recognition, and SCMA outperformed several existing methods in all aspects. In particular, the anomalies from SCMA were the most consistent with the known Fe deposits in the area, achieving an AUC of 0.89.
Cyber-attacks have become more frequent, targeted, and complex as the exponential growth in computer networks and the development of Internet of Things (IoT). Network intrusion detection system (NIDS) is an important ...
详细信息
Cyber-attacks have become more frequent, targeted, and complex as the exponential growth in computer networks and the development of Internet of Things (IoT). Network intrusion detection system (NIDS) is an important and essential tool to protect network environments. However, the low performance of a NIDS against small malicious samples has seriously threatened the security of networks, thus directly leading to the loss of personal property and national interests. Given this, we propose an auto encoder-based hybrid detection model, abbreviated as AHDM, for the intrusion detection with small-sample problem. AHDM has a dual classifier framework. It trains first neural network based on the encoding features obtained from the autoencoder feature enhancement algorithm to detect small-sample malicious traffic. It trains second neural network using the original features to detect normal traffic and large-sample malicious traffic. The final detection result of malicious traffic is obtained by combining the detection results of the two neural networks. In experiments, we use three classic datasets (KDD CUP 99, CIC-IDS-2017, and IOT-23) and simulate the malicious traffic detection targeting extremely small-sample malicious traffic. The results show that AHDM has a higher detection rate for small-sample malicious traffic compared to the advanced detection models (DNN and ACID). In the IOT-23 dataset, the AHDM model shows an absolute advantage in detecting DDoS type of malicious traffic, with a detection rate of 0.71, which is much higher than the DNN (0.14) and ACID (0.14) models.
Multimodal Emotion Recognition is challenging because of the heterogeneity gap among different modalities. Due to the powerful ability of feature abstraction, Deep Neural Networks (DNNs) have exhibited significant suc...
详细信息
Multimodal Emotion Recognition is challenging because of the heterogeneity gap among different modalities. Due to the powerful ability of feature abstraction, Deep Neural Networks (DNNs) have exhibited significant success in bridging the heterogeneity gap in cross-modal retrieval and generation tasks. In this work, a DNNs-based Multi-channel Weight-sharing autoencoder with Cascade Multi-head Attention (MCWSA-CMHA) is proposed to generically address the affective heterogeneity gap in MER. Specifically, multimodal heterogeneity features are extracted by multiple independent encoders, and then a scalable heterogeneous feature fusion module (CMHA) is realized by connecting multiple multi-head attention modules in series. The core of the proposed algorithm is to reduce the heterogeneity between the output features of different encoders through the unsupervised training of MCWSA, and then to model the affective interactions between different modal features through the supervised training of CMHA. Experimental results demonstrate that the proposed MCWSA-CMHA achieves outperformance on two publicly available datasets compared with the state-of-the-art techniques. In addition, visualization experiments and approximation experiments are used to verify the effectiveness of each module in the proposed algorithm, and the experimental results show that the proposed MCWSA-CMHA can mine more emotion-related information among multimodal features compared with other fusion methods.
This work focuses on the problem of unraveling nonlinearly mixed latent components in an unsupervised manner. The latent components are assumed to reside in the probability simplex, and are transformed by an unknown p...
详细信息
This work focuses on the problem of unraveling nonlinearly mixed latent components in an unsupervised manner. The latent components are assumed to reside in the probability simplex, and are transformed by an unknown post-nonlinear mixing system. This problem finds various applications in signal and data analytics, e.g., nonlinear hyperspectral unmixing, image embedding, and nonlinear clustering. Linear mixture learning problems are already ill-posed, as identifiability of the target latent components is hard to establish in general. With unknown nonlinearity involved, the problem is even more challenging. Prior work offered a function equation-based formulation for provable latent component identification. However, the identifiability conditions are somewhat stringent and unrealistic. In addition, the identifiability analysis is based on the infinite sample (i.e., population) case, while the understanding for practical finite sample cases has been elusive. Moreover, the algorithm in the prior work trades model expressiveness with computational convenience, which often hinders the learning performance. Our contribution is threefold. First, new identifiability conditions are derived under largely relaxed assumptions. Second, comprehensive sample complexity results are presented-which are the first of the kind. Third, a constrained autoencoder-based algorithmic framework is proposed for implementation, which effectively circumvents the challenges in the existing algorithm. Synthetic and real experiments corroborate our theoretical analyses.
This paper addresses an approach for the classification of hyperspectral imagery (HSI). In remote sensing, the HSI sensor acquires hundreds of images with narrow and continuous spectral width in visible and near-infra...
详细信息
This paper addresses an approach for the classification of hyperspectral imagery (HSI). In remote sensing, the HSI sensor acquires hundreds of images with narrow and continuous spectral width in visible and near-infrared regions of the electromagnetic (EM) spectrum. Such nature of data acquisition is very useful in the classification and/or the identification of different objects present in the HSI data. However, the low-spatial resolution and large volume of HS images make it more challenging. In the proposed approach, we use an autoencoder with convolutional neural network (AECNN) for the classification of HS images. Pre-processing with autoencoder enhances the features in the HS images which helps to obtain optimized weights in the initial layers of the CNN model. Hence, shallow CNN architecture can be utilized to extract features from the pre-processed HSI data which are used further for the classification of the same. The potential of the proposed approach has been verified by conducting many experiments on various datasets. The classification results obtained using the proposed method are compared with many state-of-the-art deep learning based methods including the winner of the geoscience and remote sensing society (GRSS) Image Fusion Contest-2018 on HSI classification held at IEEE International Geoscience and Remote Sensing Symposium (IGARSS)-2018 and it shows superiority over those methods.
Learning to discover hidden variables from unlabeled data is an important task. Traditional generative methods model the generation process of the observed variables as well as the hidden variables. However, tractable...
详细信息
Learning to discover hidden variables from unlabeled data is an important task. Traditional generative methods model the generation process of the observed variables as well as the hidden variables. However, tractable inference and learning on these models requires strong conditional independence assumptions being made among observed and hidden variables. To tackle this limitation, we propose an autoencoder framework. The encoder produces an intermediate representation from the observed variables, and the decoder is a generative latent variable model conditioned on the intermediate representation that tries to generate the hidden variables as well as to reconstruct the observed variables. We introduce three variant models of our framework with either a deterministic or a stochastic encoding process. To optimize our model, we propose an algorithm similar to the classic expectation-maximization (EM) algorithm that supports online learning for large-scale datasets. The flexibility of our framework allows us to apply it to various scenarios where the explicit inference of hidden variables is desired. We discuss the applications of our framework to the perceptual grouping task and the part-of-speech (POS) induction task. Our experiments on the two tasks demonstrate that our framework can achieve better performance than vanilla latent variable generative models.
Online reviews play an important role in influencing buyers' daily purchase decisions. However, fake and meaningless reviews, which cannot reflect users' genuine purchase experience and opinions, widely exist ...
详细信息
Online reviews play an important role in influencing buyers' daily purchase decisions. However, fake and meaningless reviews, which cannot reflect users' genuine purchase experience and opinions, widely exist on the Web and pose great challenges for users to make right choices. Therefore, it is desirable to build a fair model that evaluates the quality of products by distinguishing spamming reviews. We present an end-to-end trainable unified model to leverage the appealing properties from autoencoder and random forest. A stochastic decision tree model is implemented to guide the global parameter learning process. Extensive experiments were conducted on a large Amazon review dataset. The proposed model consistently outperforms a series of compared methods. (c) 2018 Elsevier B.V. All rights reserved.
暂无评论