检索结果-内蒙古大学图书馆

IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

作者： Fujimoto, Masakiyo Kubo, Yotaro Nakatani, Tomohiro NTT Corp NTT Commun Sci Labs Tokyo Tokyo Japan

ISBN: (纸本)9781479928934

The accurate modeling of non-stationary noise plays an important role in model-based noise suppression for noise robust speech recognition. We have already proposed methods for unsupervised noise modeling with a Gaussian mixture model or a hidden Markov model by using a minimum mean squared error estimate of the noise. However, our previous work fixed the structure of the noise model empirically without any consideration of noise characteristics;thus, optimization of the noise model structure is required if we are to obtain further improvements. Although the Bayesian information criterion (BIC) has been widely used as a conventional approach to model structure estimation, it is not always the optimal criterion. Therefore, this paper presents a way of modeling non-stationary noise with a non-parametric Bayesian approach that estimates the model structure depending on the characteristics of given observations. The proposed method provided improved results for the evaluations of two different speech recognition tasks compared with results obtained using the conventional BIC-based approach.

关键词： noise suppression unsupervised modeling non-parametric Bayesian model MMSE estimation

来源：评论

学校读者我要写书评

暂无评论

An unsupervised and Noninvasive Model for Predicting Network Resource Demands

引用

IEEE INTERNET OF THINGS JOURNAL 2018年第6期5卷 4342-4350页

作者： Corno, Fulvio De Russis, Luigi Marcelli, Andrea Montanaro, Teodoro Politecn Torino Dept Control & Comp Engn I-10129 Turin Italy

During the last decade, network providers are faced by a growing problem regarding the distribution of bandwidth and computing resources. Recently, the mobile edge computing paradigm was proposed as a possible solution, mainly in consideration of the provided possibility of transferring service demands at the edge of the network. This solution heavily relies on the dynamic allocation of resources, depending on the user needs and network connection, therefore it becomes essential to correctly predict user movements and activities. This paper proposes an unsupervised methodology to define meaningful user locations from noninvasive user information, captured by the user terminal with no computing or battery overhead. The data is analyzed through a conjoined clustering algorithm to build a stochastic Markov chain to predict the users' movements and their bandwidth demands. Such a model could be used by network operators to optimize network resources allocation. To evaluate the proposed methodology, we tested it on one of the largest public community's labeled mobile and sensor dataset, developed by the "***" initiative, and we present positive and promising results concerning the prediction capabilities of the model.

关键词： Markov chain network resources allocation noninvasive user information unsupervised modeling

来源：评论

学校读者我要写书评

暂无评论

Design of unsupervised fractional neural network model optimized with interior point algorithm for solving Bagley-Torvik equation

引用

MATHEMATICS AND COMPUTERS IN SIMULATION 2017年 132卷 139-158页

作者： Raja, Muhammad Asif Zahoor Samar, Raza Manzar, Muhammad Anwar Shah, Syed Muslim COMSATS Inst Informat Dept Elect Engn Attock Campus Attock Pakistan Capital Univ Sci & Technol Dept Elect Engn Islamabad Pakistan Int Islamic Univ Dept Elect Engn Islamabad Pakistan

In this article, an efficient computing technique has been developed for the solution of fractional order systems governed with initial value problems (IVPs) of the Bagley-Torvik equations using fractional neural networks (FNNs) optimized with interior point algorithms (IPAs). The strength of FNNs is exploited to develop an approximate model of the equation in an unsupervised manner. The training of optimal weight of the networks is carried out using IPAs. The designed scheme is evaluated on different IVPs of the equation. Comparative studies for the results of the proposed scheme are made with an available exact solution, Podlubny numerical techniques, an analytical solver based on He's variational iteration method and a reported solution of stochastic solvers based on hybrid approaches, in order to verify the correctness of the design scheme. The results of statistical analysis based on the sufficient large number of independent runs established the consistency of the proposed scheme in terms of accuracy and convergence. (C) 2016 International Association for Mathematics and Computers in Simulation (IMACS). Published by Elsevier B.V. All rights reserved.

关键词： Bagley-Torvik equation Fractional differential equations Interior point algorithm Neural networks unsupervised modeling

来源：评论

学校读者我要写书评

暂无评论

Deep variational auto-encoders for unsupervised glomerular classification

Deep variational auto-encoders for unsupervised glomerular c...

引用

SPIE Medical Imaging Symposium / 6th Digital Pathology Conference

作者： Lutnick, Brendon Yacoub, Rabi Jen, Kuang-Yu Tomaszewski, John E. Jain, Sanjay Sarder, Pinaki SUNY Buffalo Dept Pathol & Anat Sci Buffalo NY 14260 USA SUNY Buffalo Med Nephrol Buffalo NY USA Univ Calif Davis Dept Pathol Davis CA 95616 USA Washington Univ Sch Med Dept Med Nephrol St Louis MO 63130 USA

ISBN: (数字)9781510616523

ISBN: (纸本)9781510616523

The adoption of deep learning techniques in medical applications has thus far been limited by the availability of the large labeled datasets required to robustly train neural networks, as well as difficulty interpreting these networks. However, recent techniques for unsupervised training of neural networks promise to address these issues, leveraging only structure to model input data. We propose the use of a variational autoencoder (VAE) which utilizes data from an animal model to augment the training set and non-linear dimensionality reduction to map this data to human sets. This architecture utilizes variational inference, performed on latent parameters, to statistically model the probability distribution of training data in a latent feature space. We show the feasibility of VAEs, using images of mouse and human renal glomeruli from various pathological stages of diabetic nephropathy (DN), to model the progression of structural changes which occur in DN. When plotted in a 2-dimentional latent space, human and mouse glomeruli, show separation with some overlap, suggesting that the data is continuous, and can be statistically correlated. When DN stage is plotted in this latent space, trends in disease pathology are visualized.

关键词： Cross species modeling knowledge transfer variational autoencoder unsupervised modeling

来源：评论

学校读者我要写书评

暂无评论

MAPPING VERBAL ARGUMENT PREFERENCES TO DEVERBAL NOUNS

引用

INTERNATIONAL JOURNAL OF SEMANTIC COMPUTING 2009年第4期3卷 527-549页

作者： Gurevich, Olga Waterman, Scott A. Microsoft Powerset San Francisco CA 94107 USA

We describe an experiment mapping semantic role preferences for transitive verbs to their deverbal nominal forms. The preferences are learned by data mining large parsed corpora. Preferences are modeled for deverbal/argument pairs, falling back to a model for the deverbal only when sufficient data is not available. Errors in role assignment are reduced by 30-40%.

关键词： Deverbal nouns data mining unsupervised modeling parsed corpus wikipedia semantic role labeling

来源：评论

学校读者我要写书评

暂无评论

An unsupervised Framework for Online Spatiotemporal Detection of Activities of Daily Living by Hierarchical Activity Models

引用

SENSORS 2019年第19期19卷 4237-4237页

作者： Negin, Farhood Bremond, Francois INRIA 2004 Route LuciolesBP 93 F-06902 Sophia Antipolis France CNRS Inst Pascal UMR 6602 F-63171 Aubiere France Univ Cote Azur CoBTeK Team F-06108 Nice France

Automatic detection and analysis of human activities captured by various sensors (e.g., sequences of images captured by RGB camera) play an essential role in various research fields in order to understand the semantic content of a captured scene. The main focus of the earlier studies has been widely on supervised classification problem, where a label is assigned to a given short clip. Nevertheless, in real-world scenarios, such as in Activities of Daily Living (ADL), the challenge is to automatically browse long-term (days and weeks) stream of videos to identify segments with semantics corresponding to the model activities and their temporal boundaries. This paper proposes an unsupervised solution to address this problem by generating hierarchical models that combine global trajectory information with local dynamics of the human body. Global information helps in modeling the spatiotemporal evolution of long-term activities, hence, their spatial and temporal localization. Moreover, the local dynamic information incorporates complex local motion patterns of daily activities into the models. Our proposed method is evaluated using realistic datasets captured from observation rooms in hospitals and nursing homes. The experimental data on a variety of monitoring scenarios in hospital settings reveals how this framework can be exploited to provide timely diagnose and medical interventions for cognitive disorders, such as Alzheimer's disease. The obtained results show that our framework is a promising attempt capable of generating activity models without any supervision.

关键词： activity recognition activity of daily living assisted living hierarchical activity models unsupervised modeling

来源：评论

学校读者我要写书评

暂无评论

Situational Anomaly Detection in Multimedia Data under Concept Drift 21

Situational Anomaly Detection in Multimedia Data under Conce...

引用

29th ACM International Conference on Multimedia (MM)

作者： Kumari, Pratibha Indian Inst Technol Ropar Rupnagar Punjab India

ISBN: (纸本)9781450386517

Anomaly detection has been a very challenging and active area of research for decades, particularly for video surveillance. However, most of the works detect predefined anomaly classes using static models. These frameworks have limited applicability for real-life surveillance where the data have concept drift. Under concept drift, the distribution of both normal and anomaly classes changes over time. An event may change its class from anomaly to normal or vice-versa. The non-adaptive frameworks do not handle this drift. Additionally, the focus has been on detecting local anomalies, such as a region of an image. In contrast, in CCTV-based monitoring, flagging unseen anomalous situations can be of greater interest. Utilizing multiple sensory information for anomaly detection has also received less attention. This extended abstract discusses these gaps and possible solutions.

关键词： multimodal anomaly detection long term surveillance unsupervised modeling concept drift

来源：评论

学校读者我要写书评

暂无评论

Feature enhancement based on generative-discriminative hybrid approach with gmms and DNNS for noise robust speech recognition 40

Feature enhancement based on generative-discriminative hybri...

引用

40th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2015

作者： Fujimoto, Masakiyo Nakatani, Tomohiro NTT Communication Science Laboratories NTT Corporation Japan

ISBN: (纸本)9781467369978

This paper presents a technique that combines generative and discriminative approaches with Gaussian mixture models (GMMs) and deep neural networks (DNNs) for model-based feature enhancement. Typical model-based feature enhancement employs a generative model approach. The enhanced features are obtained by using the weighted sum of linear transformations given by each Gaussian component contained in GMMs and corresponding posterior probabilities. The computation of posterior probabilities is a crucial factor for this kind of feature enhancement, and can also be formulated as the class discrimination problem of observed noisy features. The prominent discriminability of DNNs is a well-known solution to this discrimination problem. Therefore, we propose the use of DNNs for computing posterior probabilities. The proposed method incorporates the benefit of the discriminative approach into the generative approach. For AURORA2 task evaluations, the proposed method provided noticeable improvements compared with results obtained using the conventional generative model approach. © 2015 IEEE.

关键词： deep neural networks feature enhancement generative-discriminative hybrid approach unsupervised modeling

来源：评论

学校读者我要写书评

暂无评论

Towards a thematic dimensional framework of online fraud: An exploration of fraudulent email attack tactics and intentions

引用

DECISION SUPPORT SYSTEMS 2023年第1期171卷

作者： Bera, Debalina Ogbanufe, Obi Kim, Dan J. Indian Inst Management Sirmaur Dept Informat Technol & Syst Sirmaur Himachal Prades India Univ North Texas G Brint Ryan Coll Business Dept Informat Technol & Decis Sci 1307 West Highland St Denton TX 76201 USA

Despite anti-phishing filters, social engineering-based cyber-attacks still result in billions of dollars lost annually, significant personal identity theft, loss of corporate secrets, and espionage. We review the phishing literature on psychological attacks and design tactics employed for deception by attackers. The result reveals the need to continuously identify tactics embedded in fraudulent email content to understand why users still fall prey to phishing attacks. Using the literature as a backdrop, we identify and conceptualize a thematic dimensional framework of fraudulent phishing attacks. This study uses benchmark datasets of fraudulent email to empirically validate the proposed dimensional framework. Specifically, a combination of machine learning-based content analysis and topic modeling found a majority of the discovered topics were labeled against the proposed dimensions. Statistical analysis was employed to provide empirical evidence of the thematic dimensions of fraudulent email attacks. This study thus, not only extends the cybersecurity literature by identifying and validating eight dimensions that enrich our understanding of phishing and attack identification but stimulates cumulative research endeavors to develop yet more comprehensive dimensional frameworks of phishing email attacks.

关键词： Spam Broad phishing Spear-phishing Thematic dimensional framework unsupervised modeling Text mining NGram Non -negative matrix factorization Topic modeling

来源：评论

学校读者我要写书评

暂无评论

Process monitoring using recurrent Kalman variational auto-encoder for general complex dynamic processes

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2023年第PartC期123卷

作者： Zhang, Zheng Zhu, Jinlin Zhang, Shuyu Gao, Furong Hong Kong Univ Sci & Technol Dept Chem & Biol Engn Kowloon Hong Kong Peoples R China Jiangnan Univ Sch Food Sci & Technol Wuxi 214122 Jiangsu Peoples R China

Recently, latent models have continued to find advantages in statistical process monitoring, especially with multifarious extensions coping with the nonlinearity and dynamics confronted by the pattern extraction. However, when later performing the famous T2 statistic on the patterns, the issue of non-Gaussianity has rarely been focused alongside, hence vitiating the control of Type-I and-II error in practice. In this paper, a model, in the name of recurrent Kalman variational auto-encoder (RKVAE), is theoretically derived for handling these properties simultaneously, and proposed for its use on monitoring general complex processes. The main idea is to build a monitoring panel on the basis of further reasoning the dynamics underlying the nonlinearly extracted Gaussian features. The model hierarchically consists of three parts: the vanilla VAE for encoding latent Gaussian features, a linear Gaussian state-space model (LGSSM) for temporally reasoning these features, and an RNN for recurrently updating the online LGSSM by fusing several candidate parameters. For the latter K in the name, it is the Kalman filter (KF) that facilitates the feature noise filtering and LGSSM identification. The model has, structurally, a good capacity to be extensible from both the depth and width when modelling complex processes. Finally, based on the explicit formula of LGSSM, a detection panel is designed, and two case studies are given to demonstrate its effectiveness in monitoring and relevant analysis.

关键词： Nonlinear process monitoring Dynamics Gaussianity Variational auto-encoder unsupervised modeling

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：