检索结果-内蒙古大学图书馆

Local Deep-Feature Alignment for Unsupervised Dimension Reduction

IEEE TRANSACTIONS ON IMAGE PROCESSING 2018年第5期27卷 2420-2432页

作者： Zhang, Jian Yu, Jun Tao, Dacheng Zhejiang Int Studies Univ Sch Sci & Technol Hangzhou 310012 Zhejiang Peoples R China Hangzhou Dianzi Univ Sch Comp Sci Hangzhou 310018 Zhejiang Peoples R China Univ Sydney UBTECH Sydney Artificial Intelligence Ctr Darlington NSW 2008 Australia Univ Sydney Sch Informat Technol Fac Engn & Informat Technol Darlington NSW 2008 Australia

This paper presents an unsupervised deep-learning framework named local deep-feature alignment (LDFA) for dimension reduction. We construct neighbourhood for each data sample and learn a local stacked contractive autoencoder (SCAE) from the neighbourhood to extract the local deep features. Next, we exploit an affine transformation to align the local deep features of each neighbourhood with the global features. Moreover, we derive an approach from LDFA to map explicitly a new data sample into the learned low-dimensional subspace. The advantage of the LDFA method is that it learns both local and global characteristics of the data sample set: the local SCAEs capture local characteristics contained in the data set, while the global alignment procedures encode the interdependencies between neighbourhoods into the final low-dimensional feature representations. Experimental results on data visualization, clustering, and classification show that the LDFA method is competitive with several well-known dimension reduction techniques, and exploiting locality in deep learning is a research topic worth further exploring.

关键词： Deep learning auto-encoder locality preserving global alignment dimension reduction

来源：评论

学校读者我要写书评

暂无评论

Gait-based Human identification using acoustic sensor and deep neural network

引用

FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE 2018年 86卷 1228-1237页

作者： Wang, Yingxue Chen, Yanan Bhuiyan, Md Zakirul Alam Han, Yu Zhao, Shenghui Li, Jianxin China Acad Elect & Informat Technol Beijing 100041 Peoples R China Beijing Inst Technol Sch Informat & Elect Beijing 100081 Peoples R China Tsinghua Univ Sch Life Sci Beijing 100084 Peoples R China Fordham Univ Dept Comp & Informat Sci Bronx NY 10458 USA Beihang Univ Sch Comp Sci & Engn Beijing 100191 Peoples R China

This paper proposes a simple, fast and low-cost gait-based human identification system jointly employing an acoustic sensor system and a deep neural network (DNN) based algorithm to process acoustic data for human identification. The acoustic sensor system is employed to detect and acquire individuals' gait signatures. Compared with the conventional radar based and video image based human gait identification technologies, our acoustic sensor system is low-cost and feasible in some special cases, such as in private rooms/hotels or in the night time. We perform identification using a DNN which consists of an auto encoder and a feed-forward neural network. The experimental results demonstrate that the integrated system of acoustic sensor and DNN considerably improves the system's performance, which achieves an excellent accuracy rate of 97% for human identification. Aside from the high accuracy of the gait-based human identification, the novel system we developed is also capable of identifying human actions (walk, jog, run, jump, etc.) and the human number which has not been addressed in the literature, thereby significantly enhancing the wide adaptability of our identification system in various applications. Additionally, to better enhance the performance of the proposed human identification method, a large feature set, variable-length frame size and variable-dimension feature vectors are exploited in our identification system. The presented work provides prospects in developing high-performance identification system for large-scale security applications via a simple and efficient technique. (C) 2017 Elsevier B.V. All rights reserved.

关键词： Acoustic sensor Human identification Deep neural network auto-encoder Feed-forward neural network

来源：评论

学校读者我要写书评

暂无评论

Applying Deep Learning for Improving Image Classification in Nuclear Fusion Devices

引用

IEEE ACCESS 2018年 6卷 72345-72356页

作者： Farias, Gonzalo Fabregas, Ernesto Dormido-Canto, Sebastian Vega, Jesus Vergara, Sebastian Dormido Bencomo, Sebastian Pastor, Ignacio Olmed, Alvaro Pontificia Univ Catolica Valparaiso Escuela Ingn Elect Ave Brasil 2147 Valparaiso 2362804 Chile Univ Nacl Educ Distancia Dept Informat & Automat E-28040 Madrid Spain CIEMAT Lab Nacl Fus E-28040 Madrid Spain

Deep learning has become one of the most promising approaches in recent years. One of the main applications of deep learning is the automatic feature extraction with auto-encoders (AEs). Feature extraction, one of the most important stages in machine learning, that can reduce drastically the dimensionality of the problem, making easier any subsequent process such as classification. The main contribution of this research is to evaluate the use of AEs for automatic feature extraction in massive thermonuclear fusion databases. In order to show the performance of AEs in a practical way, the problem of image classification of the TJ-II Thomson Scattering diagnostic has been selected. The classification has been performed by the algorithm of support vector machines and conformal predictors. The results show that the use of AEs produces the predictions faster, with more reliable models, and with higher success rates in comparison to the performance without using the deep learning approach.

关键词： Images classification auto-encoder future extraction nuclear fusion

来源：评论

学校读者我要写书评

暂无评论

Deep diagnostics and prognostics: An integrated hierarchical learning framework in PHM applications

引用

APPLIED SOFT COMPUTING 2018年 72卷 555-564页

作者： Lin, Yanhui Li, Xudong Hu, Yang Beihang Univ Sch Reliabil & Syst Engn Beijing Peoples R China Natl Space Sci Ctr Beijing Peoples R China Sci & Technol Complex Aviat Syst Simulat Lab Beijing 9236 Peoples R China

Prognostics and Health Management (PHM) is an integrated technique for improving the availability and efficiency of high-value industry equipment and reducing the maintenance cost. One of the most challenging problems in PHM is how to effectively process the raw monitoring signal into the information-rich features that are readable enough for PHM modeling. In this paper, we propose an integrated hierarchical learning framework, which is capable to perform the unsupervised feature learning, diagnostics and prognostics modeling together. The proposed method is based on auto-encoders (trained by considering the Li-norm penalty) and Extreme Learning Machines (trained by considering the L-2-norm penalty). The proposed method is applied on two different case studies considering the diagnostics of motor bearings and prognostics of turbofan engines, also the performances are compared with other commonly applied PHM approaches and machine learning tools. The obtained results demonstrate the superiority of the proposed method, especially the ability of extracting the relevant features from the non-informative and noisy signals and maintaining their efficiencies. (C) 2018 Elsevier B.V. All rights reserved.

关键词： Feature learning auto-encoder Extreme learning machines Prognostics and health management Motor bearing Turbofan engine

来源：评论

学校读者我要写书评

暂无评论

A Regularized Deep Learning Approach for Clinical Risk Prediction of Acute Coronary Syndrome Using Electronic Health Records

引用

IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING 2018年第5期65卷 956-968页

作者： Huang, Zhengxing Dong, Wei Duan, Huilong Liu, Jiquan Zhejiang Univ Coll Biomed Engn & Instrument Sci Hangzhou 310027 Peoples R China Chinese Peoples Liberat Army Gen Hosp Dept Cardiol Beijing Peoples R China

Objective: Acute coronary syndrome (ACS), as a common and severe cardiovascular disease, is a leading cause of death and the principal cause of serious long-term disability globally. Clinical risk prediction of ACS is important for early intervention and treatment. Existing ACS risk scoring models are based mainly on a small set of hand-picked risk factors and often dichotomize predictive variables to simplify the score calculation. Methods: This study develops a regularized stacked denoising autoencoder (SDAE) model to stratify clinical risks of ACS patients from a large volume of electronic health records (EHR). To capture characteristics of patients at similar risk levels, and preserve the discriminating information across different risk levels, two constraints are added on SDAE to make the reconstructed feature representations contain more risk information of patients, which contribute to a better clinical risk prediction result. Results: We validate our approach on a real clinical dataset consisting of 3464 ACS patient samples. The performance of our approach for predicting ACS risk remains robust and reaches 0.868 and 0.73 in terms of both AUC and accuracy, respectively. Conclusions: The obtained results show that the proposed approach achieves a competitive performance compared to state-of-the-art models in dealing with the clinical risk prediction problem. In addition, our approach can extract informative risk factors of ACS via a reconstructive learning strategy. Some of these extracted risk factors are not only consistent with existing medical domain knowledge, but also contain suggestive hypotheses that could be validated by further investigations in the medical domain.

关键词： Acute coronary syndrome clinical risk prediction deep learning electronic health record stacked denoising auto-encoder

来源：评论

学校读者我要写书评

暂无评论

Combining Geographical and Social Influences with Deep Learning for Personalized Point-of-Interest Recommendation

引用

JOURNAL OF MANAGEMENT INFORMATION SYSTEMS 2018年第4期35卷 1121-1153页

作者： Guo, Junpeng Zhang, Wenxiang Fan, Weiguo Li, Wenhua Tianjin Univ Dept Informat Management & Management Sci Tianjin Peoples R China Tianjin Univ Major Management Sci & Engn Tianjin Peoples R China Univ Iowa Business Analyt Iowa City IA 52242 USA Tianjin Univ Inst Syst Engn Tianjin Peoples R China

Personalized point-of-interest (POI) recommendation is important to location-based social networks (LBSNs) for helping users to explore new places and for helping third-party services to launch targeted advertisements. Discovering effective features or representations from check-in data is the key to POI recommendation. Deep learning is a representation-learning method with multiple levels for discovering intrinsic features to better represent user preferences. We analyzed users' check-in behavior in detail and developed a deep learning model to integrate geographical and social influences for POI recommendation tasks. We used a semi-restricted Boltzmann machine to model the geographical similarity and a conditional layer to model the social influence. Experiments with real-world LBSNs showed that our method performed better than other state-of-the-art methods. Theoretically, our study contributes to the effective usage of data science and analytics for social recommender system design. In practice, our results can be used to improve the quality of personalized POI recommendation services for websites and applications.

关键词： and phrases: location-based recommendations social recommendations POI recommendation deep learning auto-encoder semi-RBM Boltzmann machine geographical similarity social networks location-based commerce

来源：评论

学校读者我要写书评

暂无评论

Identification of malicious activities in industrial internet of things based on deep learning models

引用

JOURNAL OF INFORMATION SECURITY AND APPLICATIONS 2018年 41卷 1-11页

作者： AL-Hawawreh, Muna Moustafa, Nour Sitnikova, Elena Univ New South Wales ADFA Sch Engn & Informat Technol Canberra ACT Australia

Internet Industrial Control Systems (IICSs) that connect technological appliances and services with physical systems have become a new direction of research as they face different types of cyber-attacks that threaten their success in providing continuous services to organizations. Such threats cause firms to suffer financial and reputational losses and the stealing of important information. Although Network Intrusion Detection Systems (NIDSs) have been proposed to protect against them, they have the difficult task of collecting information for use in developing an intelligent NIDS which can proficiently detect existing and new attacks. In order to address this challenge, this paper proposes an anomaly detection technique for IICSs based on deep learning models that can learn and validate using information collected from TCP/IP packets. It includes a consecutive training process executed using a deep auto-encoder and deep feedforward neural network architecture which is evaluated using two well-known network datasets, namely, the NSL-KDD and UNSW-NB15. As the experimental results demonstrate that this technique can achieve a higher detection rate and lower false positive rate than eight recently developed techniques, it could be implemented in real IICS environments. (c) 2018 Elsevier Ltd. All rights reserved.

关键词： Industrial internet of things (IIoT) Internet industrial control systems (IICSs) Deep learning auto-encoder

来源：评论

学校读者我要写书评

暂无评论

Inferring contextual preferences using deep encoder-decoder learners

引用

NEW REVIEW OF HYPERMEDIA AND MULTIMEDIA 2018年第3期24卷 262-290页

作者： Unger, Moshe Shapira, Bracha Rokach, Lior Livne, Amit Ben Gurion Univ Negev Beer Sheva Israel BGU Innovat Labs Telekom Beer Sheva Israel

Context-aware systems enable the sensing and analysis of user context in order to provide personalised services. Our study is part of growing research efforts examining how high-dimensional data collected from mobile devices can be utilised to infer users' dynamic preferences that are learned over time. We suggest novel methods for inferring the category of the item liked in a specific contextual situation, by applying encoder-decoder learners (long short-term memory networks and auto encoders) on mobile sensor data. In these approaches, the encoder-decoder learners reduce the dimensionality of the contextual features to a latent representation which is learned over time. Given new contextual sensor data from a user, the latent patterns discovered from each deep learner is used to predict the liked item's category in the given context. This can greatly enhance a variety of services, such as mobile online advertising and context-aware recommender systems. We demonstrate our contribution with a point of interest (POI) recommender system in which we label contextual situations with the items' categories. Empirical results utilising a real world data set of contextual situations derived from mobile phones sensors log show a significant improvement (up to 73% improvement) in prediction accuracy compared with state of the art classification methods.

关键词： User profiling deep learning user sequence modelling auto-encoder LSTM context mobile

来源：评论

学校读者我要写书评

暂无评论

A deep cascade of neural networks for image inpainting, deblurring and denoising

引用

MULTIMEDIA TOOLS AND APPLICATIONS 2018年第22期77卷 29589-29604页

作者： Zhao, Guoping Liu, Jiajun Jiang, Jiacheng Wang, Weiying Renmin Univ China Sch Informat Beijing 100872 Peoples R China Miami Univ Dept Comp Sci & Software Engn Oxford OH 45056 USA

In recent years, we have witnessed the great success of deep learning on various problems both in low and high-level computer visions. The low-level vision problems, including inpainting, deblurring, denoising, super-resolution, and so on, are highly anticipated to occur in machine vision and image processing. Many deep learning based methods have been proposed to solve low-level vision problems. Most researches treat these problems independently;however, most of the time they appear concurrently. Motivated by the success of generative model in the field of image generation, we develop a deep cascade of neural networks to solve the inpainting, deblurring, denoising problems at the same time. Our model contains two networks: inpainting GAN and deblurring-denoising network. Inpainting GAN generates the coarse patches to fill the lost part in damaged image, and the deblurring-denoising network, stacked by a convolutional auto-encoder, will further refine them. Unlike other methods that handle each problem separately, our method jointly optimizes the two sub-networks. Because GAN training is not only unstable but also difficult, we adopt the Wasserstein distance as the loss function of the inpainting GAN and propose a gradual training strategy. Learning from the idea of residual learning, we utilize skip connections to pass image details from input to reconstruction layer. Experimental results have demonstrated that the proposed model can achieve state-of-the-art performance. Through the experiments, we also demonstrated the effectiveness of the cascade architecture.

关键词： Inpainting Deblurring Denoising Wasserstein GAN auto-encoder

来源：评论

学校读者我要写书评

暂无评论

Feature Analysis of Unsupervised Learning for Multi-task Classification Using Convolutional Neural Network

引用

NEURAL PROCESSING LETTERS 2018年第3期47卷 783-797页

作者： Kim, Jonghong Bukhari, Waqas Lee, Minho Kyungpook Natl Univ Sch Elect Engn 1370 Sankyuk Dong Taegu 702701 South Korea

This study analyzes the characteristics of unsupervised feature learning using a convolutional neural network (CNN) to investigate its efficiency for multi-task classification and compare it to supervised learning features. We keep the conventional CNN structure and introduce modifications into the convolutional auto-encoder design to accommodate a subsampling layer and make a fair comparison. Moreover, we introduce non-maximum suppression and dropout for a better feature extraction and to impose sparsity constraints. The experimental results indicate the effectiveness of our sparsity constraints. We also analyze the efficiency of unsupervised learning features using the t-SNE and variance ratio. The experimental results show that the feature representation obtained in unsupervised learning is more advantageous for multi-task learning than that obtained in supervised learning.

关键词： Unsupervised learning Convolutional neural networks Multi-task learning auto-encoder Deep learning

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：