检索结果-内蒙古大学图书馆

unsupervised neural network models of the ventral visual stream

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA 2021年第3期118卷 2014196118-2014196118页

作者： Zhuang, Chengxu Yan, Siming Nayebi, Aran Schrimpf, Martin Frank, Michael C. DiCarlo, James J. Yamins, Daniel L. K. Stanford Univ Dept Psychol Stanford CA 94305 USA Univ Texas Austin Dept Comp Sci Austin TX 78712 USA Stanford Univ Neurosci PhD Program Stanford CA 94305 USA MIT Brain & Cognit Sci 77 Massachusetts Ave Cambridge MA 02139 USA Stanford Univ Dept Comp Sci Stanford CA 94305 USA Stanford Univ Wu Tsai Neurosci Inst Stanford CA 94305 USA

Deep neural networks currently provide the best quantitative models of the response patterns of neurons throughout the primate ventral visual stream. However, such networks have remained implausible as a model of the development of the ventral stream, in part because they are trained with supervised methods requiring many more labels than are accessible to infants during development. Here, we report that recent rapid progress in unsupervised learning has largely closed this gap. We find that neural network models learned with deep unsupervised contrastive embedding methods achieve neural prediction accuracy in multiple ventral visual cortical areas that equals or exceeds that of models derived using today's best supervised methods and that the mapping of these neural network models' hidden layers is neuroanatomically consistent across the ventral stream. Strikingly, we find that these methods produce brainlike representations even when trained solely with real human child developmental data collected from head-mounted cameras, despite the fact that these datasets are noisy and limited. We also find that semisupervised deep contrastive embeddings can leverage small numbers of labeled examples to produce representations with substantially improved error-pattern consistency to human behavior. Taken together, these results illustrate a use of unsupervised learning to provide a quantitative model of a multiarea cortical brain system and present a strong candidate for a biologically plausible computational theory of primate sensory learning.

关键词： ventral visual stream deep neural networks unsupervised algorithms

来源：评论

学校读者我要写书评

暂无评论

Quantitative Comparison of unsupervised Anomaly Detection algorithms for Intrusion Detection 19

Quantitative Comparison of Unsupervised Anomaly Detection Al...

引用

34th ACM/SIGAPP Annual International Symposium on Applied Computing (SAC)

作者： Falcao, Filipe Zoppi, Tommaso Viera Silva, Caio Barbosa Santos, Anderson Fonseca, Baldoino Ceccarelli, Andrea Bondavalli, Andrea Univ Fed Alagoas Inst Comp Maceio Alagoas Brazil Dept Math & Informat Florence Italy

ISBN: (纸本)9781450359337

Anomaly detection algorithms aim at identifying unexpected fluctuations in the expected behavior of target indicators, and, when applied to intrusion detection, suspect attacks whenever the above deviations are observed. Through years, several of such algorithms have been proposed, evaluated experimentally, and analyzed in qualitative and quantitative surveys. However, the experimental comparison of a comprehensive set of algorithms for anomaly-based intrusion detection against a comprehensive set of attacks datasets and attack types was not investigated yet. To fill such gap, in this paper we experimentally evaluate a pool of twelve unsupervised anomaly detection algorithms on five attacks datasets. Results allow elaborating on a wide range of arguments, from the behavior of the individual algorithm to the suitability of the datasets to anomaly detection. We identify the families of algorithms that are more effective for intrusion detection, and the families that are more robust to the choice of configuration parameters. Further, we confirm experimentally that attacks with unstable and non-repeatable behavior are more difficult to detect, and that datasets where anomalies are rare events usually result in better detection scores.

关键词： Anomaly Detection Intrusion Detection unsupervised algorithms Comparison Attacks Datasets Attack Model

来源：评论

学校读者我要写书评

暂无评论

YAC2: An α-proximity based clustering algorithm

引用

EXPERT SYSTEMS WITH APPLICATIONS 2021年 167卷 114138-114138页

作者： Ghiassi, M. Saidane, H. Oswal, R. Santa Clara Univ 316PLucas Hall500 El Camino Real Santa Clara CA 95053 USA Optimal Data Sci Inc 2025 Portsmouth Dr Richardson TX 75082 USA Santa Clara Univ 500 Camino Real Santa Clara CA 95053 USA

Clustering is the process of identifying objects with similar intrinsic properties and grouping them together into separate clusters. It also serves as a preliminary step in data classification by separating heterogeneous data into reasonably homogeneous groups, which can be further processed. Existing literature discusses a host of different clustering algorithms and their applications, albeit no single approach for all applications has yet emerged. In this paper, we introduce a novel clustering algorithm, YAC2, based on data binning and alpha-proximity/neighborhood. The binning process is a data transformation step that converts cardinal values of the attributes into their ordinal equivalence. The algorithm introduces a specialized centroid that is used with the alpha-proximity and a matching algorithm to partition data set into a sequentially generated clusters. We present the results of applying YAC2 to a set of established benchmark data sets, using a host of evaluation metrics. Our results show YAC2 to perform well besting a number of well-established algorithms. For several metrics, YAC2 has provided improvements averagely in the range of 6% (in the Iris data set)-180% (in the Wholesale Customer data set) over other algorithms.

关键词： Clustering analysis YAC2 Machine learning unsupervised algorithms

来源：评论

学校读者我要写书评

暂无评论

Aligning Sentences Between Comparable Texts of Different Styles 1

引用

9th Joint International Semantic Technology Conference (JIST)

作者： Chen, Xiwen Zhang, Mengxue Zhu, Kenny Qili Shanghai Jiao Tong Univ Comp Sci Dept Adv Data & Programming Technol Lab 800 Dong Chuan Rd Shanghai Peoples R China

ISBN: (数字)9789811534126

ISBN: (纸本)9789811534126;9789811534119

Monolingual parallel corpus is crucial for training and evaluating text rewriting or paraphrasing models. Aligning parallel sentences between two large body of texts is a key step toward automatic construction of such parallel corpora. We propose a greedy alignment algorithm that makes use of strong unsupervised similarity measures. The algorithm aligns sentences with state-of-the-art accuracy while being more robust on corpora with special linguistic features. Using this alignment algorithm, we automatically constructed a large English parallel corpus from various translated works of classic literature.

关键词： Monolingual parallel corpora Sentence alignment unsupervised algorithms

来源：评论

学校读者我要写书评

暂无评论

unsupervised change detection technique based on fully convolutional network using RGBD

Unsupervised change detection technique based on fully convo...

引用

作者： Yan, Jianda University of Twente

学位级别：硕士

Due to the rapid process of urbanization, new build up areas and infrastructures like buildings, streets, bridges, and other man-made objects are changing the world all the time. Therefore, there is an increasing demand for detecting changes in urban areas. Traditional two-dimensional (2D) change detection methods are limited by different image perspective and illumination. This thesis describes a 3D change detection methodology by the joint use of height and spectral information. The proposed method follows the following steps. Firstly, the subtraction of DSMs and a morphological based post-processing are performed between image pairs. Then, the effect of several RGB-based methods is compared and analyzed based on the study area. Furthermore, we combine the DSM-based method and RGB-based. Finally, we define and calculate the 'reliability' for the labels obtained from the combined method, and selecting reliable labels as training samples according to the reliability to train the FCN network. This method enables the FCN architecture to work without manually labeled training samples, which is a high labor cost and is a time- consuming task. By using the FCN architecture, additional contextual information can be considered, and results derived from the joint of DSM-based and RGB-based method can be further improved. After checking the result, we find that the errors caused by shadows, different seasons, and the growth of vegetation are reduced. Also, the noises generated in the pixel-based method are also removed by the FCN architecture. The study area is the city of Ecublens, Switzerland. The data is orthophoto acquired by the UAV, and the DSM data is generated from photogrammetry using the overlapping images. Images are obtained from three different times, and all experiments are performed three times, each time taking two images obtained from different times for change detection. In the end, our method is compared to the supervised FCN architecture, which uses manua

关键词： Change detection method unsupervised algorithms Fully convolutional neural network RGBD data

来源：评论

学校读者我要写书评

暂无评论

Nonlinear projection methods for visualizing Barcode data and application on two data sets

引用

MOLECULAR ECOLOGY RESOURCES 2013年第6期13卷 976-990页

作者： Olteanu, Madalina Nicolas, Violaine Schaeffer, Brigitte Denys, Christiane Missoup, Alain-Didier Kennis, Jan Laredo, Catherine Univ Paris 01 EA 4543 F-75013 Paris France INRA Math & Informat Appl UR341 F-78350 Jouy En Josas France Museum Natl Hist Nat CNRS UMR 7205 Dept Systemat & Evolut F-75005 Paris France Univ Douala Fac Sci Dept Anim Biol Organisms Douala Cameroon Univ Antwerp Evolutionary Ecol Grp B-2020 Antwerp Belgium Univ Paris Diderot CNRS UMR 7599 LPMA F-75013 Paris France

Developing tools for visualizing DNA sequences is an important issue in the Barcoding context. Visualizing Barcode data can be put in a purely statistical context, unsupervised learning. Clustering methods combined with projection methods have two closely linked objectives, visualizing and finding structure in the data. Multidimensional scaling (MDS) and Self-organizing maps (SOM) are unsupervised statistical tools for data visualization. Both algorithms map data onto a lower dimensional manifold: MDS looks for a projection that best preserves pairwise distances while SOM preserves the topology of the data. Both algorithms were initially developed for Euclidean data and the conditions necessary to their good implementation were not satisfied for Barcode data. We developed a workflow consisting in four steps: collapse data into distinct sequences;compute a dissimilarity matrix;run a modified version of SOM for dissimilarity matrices to structure the data and reduce dimensionality;project the results using MDS. This methodology was applied to Astraptes fulgerator and Hylomyscus, an African rodent with debated taxonomy. We obtained very good results for both data sets. The results were robust against unbalanced species. All the species in Astraptes were well displayed in very distinct groups in the various visualizations, except for LOHAMP and FABOV that were mixed up. For Hylomyscus, our findings were consistent with known species, confirmed the existence of four unnamed taxa and suggested the existence of potentially new species.

关键词： Barcode dissimilarity matrices DNA sequences median self-organizing maps multidimensional scaling unsupervised algorithms visualization

来源：评论

学校读者我要写书评

暂无评论

unsupervised Graph-Based Similarity Learning Using Heterogeneous Features

Unsupervised Graph-Based Similarity Learning Using Heterogen...

引用

作者： Muthukrishnan, Pradeep University of Michigan

学位级别：Ph.D.

Relational data refers to data that contains explicit relations among objects. Nowadays, relational data are universal and have a broad appeal in many different application domains. The problem of estimating similarity between objects is a core requirement for many standard Machine Learning (ML). Natural Language Processing (NLP) and Information Retrieval (IR) problems such as clustering, classification, word sense disambiguation, etc. Traditional machine learning approaches represent the data using simple, concise representations such as feature vectors. While this works very well for homogeneous data, i.e., data with a single feature type such as text, it does not exploit the availability of different feature types fully. For example, scientific publications have text, citations, authorship information, venue information. Each of the features can be used for estimating similarity. Representing such objects has been a key issue in efficient mining (Getoor and Taskar, 2007). In this thesis, we propose natural representations for relational data using multiple, connected layers of graphs; one for each feature type. Also, we propose novel algorithms for estimating similarity using multiple heterogeneous features. Also, we present novel algorithms for tasks like topic detection and music recommendation using the estimated similarity measure. We demonstrate superior performance of the proposed algorithms (root mean squared error of 24.81 on the Yahoo! KDD Music recommendation data set and classification accuracy of 88% on the ACL Anthology Network data set) over many of the state of the art algorithms, such as Latent Semantic Analysis (LSA), Multiple Kernel Learning (MKL) and spectral clustering and baselines on large, standard data sets.

关键词： unsupervised algorithms Graph-based Learning Similarity Learning Machine Learning Multiple Heterogeneous Features Thesis

来源：评论

学校读者我要写书评

暂无评论

An MSE-Based Method to Avoid Permutation/Gain Indeterminacy in Frequency-Domain Blind Source Separation

引用

CIRCUITS SYSTEMS AND SIGNAL PROCESSING 2010年第3期29卷 403-417页

作者： Dapena, Adriana Iglesia, Daniel Escudero, Carlos J. Univ A Coruna Fac Informat Dept Elect & Sistemas Corunna 15071 Spain

This paper presents a novel technique for separating convolutive mixtures of statistically independent non-Gaussian signals without resorting to an a priori knowledge of the sources or the mixing system. This problem is solved in the frequency domain by transforming the convolutive mixture into several instantaneous mixtures which are independently separated using blind source separation (BSS) algorithms. First, the instantaneous mixture at one frequency is solved using the joint approximate diagonalization of eigenmatrices (JADE) technique, and the other mixtures are then separated using the mean squared error (MSE) criterion. As a special case of this method, we consider the separation of non-Gaussian temporally white signals transmitted in blocks with zero padding between them.

关键词： Convolutive mixtures Blind source separation JADE Frequency-domain systems unsupervised algorithms

来源：评论

学校读者我要写书评

暂无评论

Mixed Group Discovery: Incorporating Group Linkage with Alternatively Consistent Social Network Analysis

Mixed Group Discovery: Incorporating Group Linkage with Alte...

引用

4th Annual IEEE International Conference on Semantic Computing (ICSC)

作者： Huang, Shu Penn State Univ Coll Informat Sci & Technol Univ Pk State Coll PA 16801 USA

ISBN: (纸本)9780769541549

Poor quality data exists widely in various database applications. In relational database, each entity is associated with a group of relational records. For two entities with similar identifiers, the records of one entity may be mistakenly combined into the group of the other. This is referred to as the mixed group problem. In this paper, we formulate the problem of discovering mixed groups and propose two unsupervised algorithms as solution. From the relational records, we observe that a group in one database is unlikely to be mixed in the same pattern as that of the same entity in another independent database. Also we find that the collaborative relationship between entities tend to be alternatively consistent over time. By investigating and applying these properties, we propose two mixed group discovery algorithms, as well as a generic model that covers various situations. Empirical experiments on both synthetic and real datasets from Citeseer and ACM digital libraries show that our algorithms can identify mixed groups with more than 70% precision and 80% recall, and the overall performance is significantly better than existing methods.

关键词： ACM digital libraries Citeseer digital libraries digital libraries group linkage independent database mixed group discovery algorithms relational database relational databases social network analysis social networking (online) unsupervised algorithms unsupervised 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：