检索结果-内蒙古大学图书馆

autoencoder embedded dictionary learning for nonlinear industrial process fault diagnosis

JOURNAL OF PROCESS CONTROL 2021年 101卷 24-34页

作者： Li, Yanxia Chai, Yi Yin, Hongpeng Chongqing Univ Coll Automat Chongqing 400044 Peoples R China State Key Lab Power Transmiss Equipment & Syst Se Chongqing 400030 Peoples R China

Industrial processes usually exhibit great nonlinearity generated from the effects of complex mechanisms, system integrations and multiple working conditions. Although a variety of dictionary learning algorithms have been proposed in recent years for industrial process fault diagnosis, most of them only model the process data via a linear combination of a few dictionary atoms, which cannot effectively characterize the nonlinear relationships among variables and may lead to limited diagnosis performance. Recent improvements in multilayer neural networks, especially the autoencoders, offer opportunities to tackle the nonlinear problem. However, the overall limited availability of fault samples poses great challenges in achieving satisfactory performance. To address the mentioned issues simultaneously, the present study proposes an autoencoder Embedded Dictionary Learning approach (AEDL) for nonlinear industrial process fault diagnosis. First, an autoencoder is employed to learn a nonlinear mapping that maps the linearly inseparable industrial process data to a high-dimensional space, where a desired dictionary is learned according to the basic dictionary learning algorithm. Next, two supervised graphs, leveraging the priors of industrial process data, are introduced into the learning process to make the proposed approach robust to training samples. After obtaining the dictionary, the coding coefficients of the process data over the dictionary can be used for fault diagnosis via a simple classifier. As revealed from the encouraging experimental results on the Tennessee Eastman process, the developed approach outperforms several dictionary learning approaches and some other nonlinear fault diagnosis methods. (C) 2021 Published by Elsevier Ltd.

关键词： Fault diagnosis Nonlinear industrial process Dictionary learning autoencoder

来源：评论

学校读者我要写书评

暂无评论

autoencoder framework based on orthogonal projection constraints improves anomalies detection

引用

NEUROCOMPUTING 2021年 450卷 372-388页

作者： Yu, Qien Kavitha, Muthusubash Kurita, Takio Hiroshima Univ Dept Informat Engn Higashihiroshima Hiroshima 7398521 Japan Hiroshima Univ Grad Sch Adv Sci & Engn Higashihiroshima Hiroshima 7398521 Japan Nagasaki Univ Sch Informat & Data Sci Nagasaki Japan

In this study, we propose a novel autoencoder framework based on orthogonal projection constraint (OPC) for anomaly detection (AD) on both complex image and vector datasets. Orthogonal projection is useful to capture the null subspace that consists of noisy information for AD, which is explicitly ignored in the existing approaches. The exploration of double subspaces, called normal space (NS) and abnormal space (AS) can improve the discriminative manifold information. Therefore, in this study, autoencoder framework based on the OPC learning method is proposed that combines the orthogonal subspace score and the reconstruction error score in the target tasks for AD. To the best of our knowledge, this is the first study that introduces an autoencoder-based model with two orthogonal subspaces for AD. Through the orthogonality, the anomaly-free data and abnormalnnosiy information are projected into the NS and the AS, respectively. Thus, it potentially addresses the problem of the distribution of generative model by combining the abilities of two subspaces that can appropriately learn the features and establish a strict boundaries around the normal data. For image datasets, we propose a convolutional autoencoder based on OPC. Additionally, the generalization and adaptability of the proposed method in AD was investigated using vector datasets by implementing a fully-connected layer-based OPC in the encoder-decoder structure. The effectiveness of the proposed framework for AD was evaluated through the comparison with state-of-the-art approaches. (c) 2021 Elsevier B.V. All rights reserved.

关键词： Orthogonal projection autoencoder Anomaly detection Subspace detection

来源：评论

学校读者我要写书评

暂无评论

autoencoder-based Anomaly Detection for Time Series Data in Complex Systems

Autoencoder-based Anomaly Detection for Time Series Data in ...

引用

IEEE Asia Pacific Conference on Circuits and Systems (APCCAS)

作者： Gong, Xundong Liao, Shibo Hu, Fei Hu, Xiaoqing Liu, Chunshan State Grid Jiangsu Elect Power Co Nanjing Peoples R China Hangzhou Dianzi Univ Hangzhou Peoples R China State Grid Wuxi Power Supply Co Wuxi Peoples R China

ISBN: (纸本)9781665450737

In this paper, we present a new anomaly detection method for time-series data in complex systems such as power grid and cellular networks. The proposed anomaly detection method is developed following unsupervised learning, where an autoencoder based on Gated Recurrent Units (GRU-AE) is trained to reconstruct a time-series of interest, and anomalies are detected via detecting exceptionally large reconstruction errors. A multi-timestamp stacking method is adopted to reduce the number of time steps in the GRU-AE to facilitate the training of the model and a new training scheme with random shuffling is proposed to prevent overfitting. The proposed GRU-AE based detector is applied in multiple time scales to detect different types of anomalies. Numerical results obtained via time-series data from real cellular network demonstrate the performance of the proposed method.

关键词： time series anomaly detection autoencoder multi-timestamp stacking random shuffling

来源：评论

学校读者我要写书评

暂无评论

Comparison of autoencoder Models for Unsupervised Representation Learning of Skeleton Sequences 18th

Comparison of Autoencoder Models for Unsupervised Representa...

引用

18th International Symposium on Visual Computing (ISVC)

作者： Zaw, Thura Zhou, Jiaxin Komuro, Takashi Saitama Univ Grad Sch Sci & Engn Saitama 3388570 Japan

ISBN: (纸本)9783031479687;9783031479694

This study comprehensively explores the performance of six distinct feature extraction models on two widely used skeleton datasets, UCLA and NTU 60, with a focus on establishing the superiority of hybrid transformer architectures. The models include autoencoder, convolutional autoencoder, temporal autoencoder (GRU autoencoder), transformer, convolutional transformer, and transformer with a convolutional encoder, representing diverse approaches in skeleton data analysis. The primary objective is to compare the models' effectiveness in extracting meaningful features from the skeleton datasets. Rigorous evaluations and comparisons are conducted using quantitative measures and visualizations to assess each model's discriminative power in capturing relevant information. The findings demonstrate that the transformer model consistently outperforms all other models on both datasets, showcasing its unique capability in extracting meaningful and discriminative features from skeleton data. Additionally, the study investigates hybrid models, combining transformer architectures with other feature extraction techniques, revealing their potential to surpass individual model capabilities. These findings can guide future studies in selecting appropriate models for similar tasks and promoting the development of more accurate and robust action recognition systems based on skeleton data.

关键词： Human Action Recognition (HAR) autoencoder Transformer Convolution Features extraction

来源：评论

学校读者我要写书评

暂无评论

autoencoder for Semisupervised Multiple Emotion Detection of Conversation Transcripts

引用

IEEE TRANSACTIONS ON AFFECTIVE COMPUTING 2021年第3期12卷 682-691页

作者： Duc-Anh Phan Matsumoto, Yuji Shindo, Hiroyuki Nara Inst Sci & Technol Nara 6300192 Japan Nara Inst Sci & Technol Informat Sci Nara 6300192 Japan

Textual emotion detection is a challenge in computational linguistics and affective computing study as it involves the discovery of all associated emotions expressed within a given piece of text. It becomes an even more difficult problem when applied to conversation transcripts, as we need to model the spoken utterances between speakers, keeping in mind the context of the entire conversation. In this paper, we propose a semisupervised multilabel method of predicting emotions from conversation transcripts. The corpus contains conversational quotes extracted from movies. A small number of them are annotated, while the rest are used for unsupervised training. We use the word2vec word-embedding method to build an emotion lexicon from the corpus and to embed the utterances into vector representations. A deep-learning autoencoder is then used to discover the underlying structure of the unsupervised data. We fine-tune the learned model on labeled training data, and measure its performance on a test set. The experiment result suggests that the method is effective and is only slightly behind human annotators.

关键词： Motion pictures Correlation Social network services Neural networks Context modeling Data models Training data Emotion recognition semisupervised learning multilabel word2vec autoencoder

来源：评论

学校读者我要写书评

暂无评论

autoencoder-based deep metric learning for network intrusion detection

引用

INFORMATION SCIENCES 2021年 569卷 706-727页

作者： Andresini, Giuseppina Appice, Annalisa Malerba, Donato Univ Bari Aldo Moro Dept Informat Via Orabona 4 I-70125 Bari Italy Consorzio Interuniv Nazl Infomat CINI Rome Italy

Nowadays intrusion detection systems are a mandatory weapon in the war against the ever-increasing amount of network cyber attacks. In this study we illustrate a new intrusion detection method that analyses the flow-based characteristics of the network traffic data. It learns an intrusion detection model by leveraging a deep metric learning methodology that originally combines autoencoders and Triplet networks. In the training stage, two separate autoencoders are trained on historical normal network flows and attacks, respectively. Then a Triplet network is trained to learn the embedding of the feature vector representation of network flows. This embedding moves each flow close to its reconstruction, restored with the autoencoder associated with the same class as the flow, and away from its reconstruction, restored with the autoencoder of the opposite class. The predictive stage assigns each new flow to the class associated with the autoencoder that restores the closest reconstruction of the flow in the embedding space. In this way, the predictive stage takes advantage of the embedding learned in the training stage, achieving a good prediction performance in the detection of new signs of malicious activities in the network traffic. In fact, the proposed methodology leads to better predictive accuracy when compared to competitive intrusion detection architectures on benchmark datasets. (c) 2021 Elsevier Inc. All rights reserved.

关键词： Network intrusion detection Deep metric learning Triplet network autoencoder

来源：评论

学校读者我要写书评

暂无评论

autoencoder-bank based design for adaptive channel-blind robust transmission

引用

EURASIP JOURNAL ON WIRELESS COMMUNICATIONS AND NETWORKING 2021年第1期2021卷 1-15页

作者： Safi, Hossein Akbari, Mohammad Vaezpour, Elaheh Parsaeefard, Saeedeh Shubair, Raed M. Shahid Beheshti Univ Dept Elect Engn Tehran Iran ICT Res Inst ITRC Dept Commun Technol Tehran Iran Univ Toronto Dept Elect & Comp Engn Toronto ON Canada MIT 77 Massachusetts Ave Cambridge MA 02139 USA New York Univ NYU Abu Dhabi Abu Dhabi 129188 U Arab Emirates

The idea of employing deep autoencoders (AEs) has been recently proposed to capture the end-to-end performance in the physical layer of communication systems. However, most of the current methods for applying AEs are developed based on the assumption that there exists an explicit channel model for training that matches the actual channel model in the online transmission. The variation of the actual channel indeed imposes a major limitation on employing AE-based systems. In this paper, without relying on an explicit channel model, we propose an adaptive scheme to increase the reliability of an AE-based communication system over different channel conditions. Specifically, we partition channel coefficient values into sub-intervals, train an AE for each partition in the offline phase, and constitute a bank of AEs. Then, based on the actual channel condition in the online phase and the average block error rate (BLER), the optimal pair of encoder and decoder is selected for data transmission. To gain knowledge about the actual channel conditions, we assume a realistic scenario in which the instantaneous channel is not known, and propose to blindly estimate it at the Rx, i.e., without any pilot symbols. Our simulation results confirm the superiority of the proposed adaptive scheme over existing methods in terms of the average power consumption. For instance, when the target average BLER is equal to 10-4, our proposed algorithm with 5 pairs of AE can achieve a performance gain over 1.2 dB compared with a non-adaptive scheme.

关键词： autoencoder Blind estimation Deep learning

来源：评论

学校读者我要写书评

暂无评论

Fusing separated representation into an autoencoder for magnetic materials outlier detection

引用

SYSTEMS SCIENCE & CONTROL ENGINEERING 2022年第1期10卷 181-191页

作者： Cao, Ying Ko, SunWoo Jiangxi Univ Sci & Technol Informat Ctr Ganzhou Peoples R China Jeonju Univ Dept Artificial Intelligence Jeonju South Korea

In materials science, an outlier may be due to variability in measurement, or it may indicate experimental errors. In this paper, we used an unsupervised method to remove outliers before further data-driven material analysis. Recently, autoencoder networks have achieved excellent results by minimizing reconstruction error. However, autoencoders do not promote the separation between outliers and inliers. The proposed SRAE model integrates latent representation to optimize the reconstruction error and ensures that outliers always deviate from the dataset in the compressed representation space. Experiments on the Nd-Fe-B magnetic materials dataset also show that after removing outliers with the proposed method, the prediction result of material property is significantly improved, indicating that the outlier detection effect is excellent.

关键词： autoencoder outlier detection latent representation outlier score magnetic materials

来源：评论

学校读者我要写书评

暂无评论

Enhancing Traffic Data Prediction: Fusion of autoencoder-Based Spatiotemporal Reconstruction and Forecasting 3

Enhancing Traffic Data Prediction: Fusion of Autoencoder-Bas...

引用

3rd International Conference on Big Data, Information and Computer Network, BDICN 2024

作者： Gao, Min Wei, Yingmei College of Systems Engineering National University of Defense Technology Changsha China

ISBN: (纸本)9798350363821

Urban traffic prediction plays a crucial role in modern city planning and management. Accurate traffic forecasts are essential for optimizing transportation systems, reducing congestion, and enhancing overall urban mobility. In this paper, we introduce a groundbreaking approach to traffic data prediction, leveraging the power of autoencoders and a reimagined training methodology. The conventional methods, while effective to some extent, are inherently limited by their exclusive reliance on a fixed window of preceding steps. This limitation impedes the ability of these methods to adequately consider the continuity between the content to be predicted and the preceding context, particularly when dealing with dynamic and evolving patterns inherent in grid-based traffic data. In recognition of these shortcomings, our method introduces a paradigm shift by extending the prediction horizon and employing an autoencoder for sequence reconstruction. This approach introduces a dual loss paradigm, encompassing both reconstruction and prediction losses. To evaluate our method, we conducted experiments on two grid-based traffic datasets, demonstrating remarkable improvements in prediction accuracy. Our approach outperforms the current methods, highlighting its potential for advancing urban traffic prediction and contributing to more efficient and sustainable transportation systems. The results underscore the promising future of incorporating autoencoder-driven spatiotemporal information for urban traffic forecasting, contributing insights towards a novel approach for city traffic management and planning. © 2024 IEEE.

关键词： autoencoder reconstruction loss spatiotemporal data traffic prediction

来源：评论

学校读者我要写书评

暂无评论

autoencoder networks extract latent variables and encode these variables in their connectomes

引用

NEURAL NETWORKS 2021年 141卷 330-343页

作者： Farrell, Matthew Recanatesi, Stefano Mihalas, Stefan Reid, R. Clay Shea-Brown, Eric Univ Washington Appl Math Dept Seattle WA 98195 USA Univ Washington Computat Neurosci Ctr Seattle WA 98195 USA Allen Inst Brain Sci Seattle WA USA

Advances in electron microscopy and data processing techniques are leading to increasingly large and complete microscale connectomes. At the same time, advances in artificial neural networks have produced model systems that perform comparably rich computations with perfectly specified connectivity. This raises an exciting scientific opportunity for the study of both biological and artificial neural networks: to infer the underlying circuit function from the structure of its connectivity. A potential roadblock, however, is that - even with well constrained neural dynamics - there are in principle many different connectomes that could support a given computation. Here, we define a tractable setting in which the problem of inferring circuit function from circuit connectivity can be analyzed in detail: the function of input compression and reconstruction, in an autoencoder network with a single hidden layer. Here, in general there is substantial ambiguity in the weights that can produce the same circuit function, because largely arbitrary changes to input weights can be undone by applying the inverse modifications to the output weights. However, we use mathematical arguments and simulations to show that adding simple, biologically motivated regularization of connectivity resolves this ambiguity in an interesting way: weights are constrained such that the latent variable structure underlying the inputs can be extracted from the weights by using nonlinear dimensionality reduction methods. (C) 2021 Elsevier Ltd. All rights reserved.

关键词： autoencoder Connectome Dimensionality reduction Identifiability Latent variable encoding

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：