检索结果-内蒙古大学图书馆

Probabilistic based recursive model for adaptive processing of data structures

EXPERT SYSTEMS WITH applications 2008年第2期34卷 1403-1422页

作者： Cho, Siu-Yeung Nanyang Technol Univ Sch Comp Engn Div Comp Syst Singapore 639798 Singapore

One of the most popular frameworks for the adaptive processing of data structures to date, was proposed by Frasconi et al. [Frasconi, P., Gori, M., & Sperduti, A. (1998). A general framework for adaptive processing of data structures. IEEE Transactions oil Neural Networks, 9(September), 768-785], who used a Backpropagation Through Structures (BPTS) algorithm [Goller, C., & Kuchler, A. (1996). Learning task-dependent distributed representations by back-propagation through structures. In Proceedings of IEEE international conference on neural networks (pp. 347-352);Tsoi, A. C. (1998). Adaptive processing of data structure: An expository overview and comments. Technical report in Faculty Informatics. Wollongong, Australia: University of Wollongong] to carry out supervised learning. This supervised model has been successfully applied to a number of learning tasks that involve complex symbolic structural patterns, such as image semantic structures, internet behavior, and chemical compounds. In this paper, we extend this model, using probabilistic estimates to acquire discriminative information from the learning patterns. Using this probabilistic estimation, smooth discriminant boundaries can be obtained through a process of clustering onto the observed input attributes. This approach enhances the ability of class discrimination techniques to recognize structural patterns. The proposed model is represented by a set of Gaussian Mixture Models (GMMs) at the hidden layer and a set of "weighted sum input to sigmoid function" models at the output layer. The proposed model's learning framework is divided into two phases: (a) locally unsupervised learning for estimating the parameters of the GMMs and (b) globally supervised learning for fine-tuning the GMMs' parameters and optimizing weights at the output layer. The unsupervised learning phase is formulated as a maximum likelihood problem that is solved by the expectation-maximization (EM) algorithm. The supervised learning phase

关键词： adaptive processing of data structures probabilistic recursive model Gaussian mixture model expectation-maximization algorithm Levenberg-Marquardt algorithm

来源：评论

学校读者我要写书评

暂无评论

Biometric Security using EEG Signal processing - Acquisition, Representation and Classification Approaches

Biometric Security using EEG Signal Processing - Acquisition...

引用

2022 IEEE international conference on distributed Computing and Electrical Circuits and Electronics, ICDCECE 2022

作者： Biradar, Shashank D. Nalbalwar, Sanjay L. Deosarkar, Shankar B. Dbatu Dept. of E&tc Engineering 402104 India Vpkbiet 413133 India

ISBN: (纸本)9781665483162

Authentication and security issues are rapidly increasing with the development of new coding techniques and network interfacing. Biometric security has shown a promising performance for security mechanism compared to key based security mechanisms. Biometric security system operates on physiological and neurological inputs and their behavioral characteristic. In Biometric security system, forging and synthetic false inputs limits its application and reduces the reliability. Inputs such as face, iris, voice etc., are highly vulnerable to such attack. Biological signaling observes a higher reliability in providing security. Electroencephalogram (EEG) has gained a large interest in recent past for biometric security provisioning. This paper outlines a brief review on the recent development in the area of provisioning security using EEG signal for biometric security. Various approaches of EEG interface for biometric security, classification and its applications were reviewed. © 2022 IEEE.

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

distributed Training Based on Horizontal Scaling for Food Image Classification 4

Distributed Training Based on Horizontal Scaling for Food Im...

引用

4th international conference on Embedded and distributed Systems, EDiS 2024

作者： El Amin, Gasbaoui Mohammed Soumia, Benkrama Mostefa, Bendjima Sofiane, Abden Ikram, Kourtiche Tahri Mohammed University Laboratory of Tit Mathematics and Computer Science Department Bechar Algeria

ISBN: (纸本)9798331519872

Deep learning models for food image classification rely on vast amounts of data to effectively recognize and differentiate between various food items. However, training these models on such extensive datasets presents significant computational challenges. Big data and distributed training are inherently complementary in addressing these challenges. Big data involves large volumes of information that traditional single-machine processing cannot handle efficiently. distributed training offers a solution by enabling machine learning models to scale across multiple machines or nodes, thereby more effectively processing and learning from extensive datasets. By partitioning tasks into smaller chunks processed in parallel, distributed training optimizes computational resources, ensuring efficient analysis of big data while maintaining fault tolerance and performance scalability. This paper explores various strategies for distributed training and communication between GPUs. We compare the performance and utilization of a single P100 GPU versus two T4 GPUs on the Food-11 dataset, focusing on accuracy and training time. Our results indicate that using two GPUs yields slightly better performance than a single GPU, reducing the training time by 125 seconds, decreasing the validation dataset loss by 0.02, and increasing the accuracy by 1%. The overall accuracy achieved is 92%, with a corresponding loss of 0.28. © 2024 IEEE.

关键词： big data data distribution deep learning food images GPUs parallel processing PyTorch

来源：评论

学校读者我要写书评

暂无评论

Research on the Key Technology of Big Data Service in University Library 13

Research on the Key Technology of Big Data Service in Univer...

引用

13th international conference on Natural Computation, Fuzzy Systems and Knowledge Discovery (ICNC-FSKD)

作者： Ye, Chunlei Lib Beijing Univ Agr Informat Consultat Dept Beijing Peoples R China

ISBN: (纸本)9781538621653

With the development of information technology of university library, the mass data of the university library has the basic characteristics of Big Data. However, the current situation of the university library is the lack of distributed storage and computing model for massive data, the lack of capacity to handl the diverse data sources, including the structured, semi structured and unstructured data, the lack of a simple, flexible application model of big data *** order to solve the problems in the service innovation of University Libraries in China, such as the problem of distributed storage and computation of massive data, the distributed management of diverse data sources, the simple and flexible application of big data services, this paper analyzes the research contents of big data processing, Hadoop ecosystem and the demand for big data services in University Libraries, and presents a technology framework for big data service in University Libraries based Hadoop. The framework includes the distributed storage and parallel computing model of mass data, the distributed management model of diverse data sources and the model of diversified service application for university libraries. This framework takes full account of the service innovation change of University Library under the environment of big data, such as data storage and calculation, data management and service applications et al. It can solve the key technical problems of big data service of University Library in a certain extent.

关键词： Hadoop Big Data hadoop big data service University Libraries

来源：评论

学校读者我要写书评

暂无评论

Online Automated Reliability Classification of Queueing Models for Streaming processing Using Support Vector Machines 21st

Online Automated Reliability Classification of Queueing Mode...

引用

21st international conference on parallel and distributed Computing (Euro-Par)

作者： Beard, Jonathan C. Epstein, Cooper Chamberlain, Roger D. Washington Univ Dept Comp Sci & Engn St Louis MO 63130 USA

ISBN: (纸本)9783662480960;9783662480953

When do you trust a performance model? More specifically, when can a particular model be used for a specific application? Once a stochastic model is selected, its parameters must be determined. This involves instrumentation, data collection, and finally interpretation;which are very time consuming. Even when done correctly, the results hold for only the conditions under which the system was characterized. For modern, dynamic stream processing systems, this is far too slow if a model-based approach to performance tuning is to be considered. This work demonstrates the use of a Support Vector Machine (SVM) to determine if a stochastic queueing model is usable or not for a particular queueing station within a streaming application. When combined with methods for online service rate approximation, our SVM approach can select models while the application is executing (online). The method is tested on a variety of hardware and software platforms. The technique is shown to be highly effective for determining the applicability of M/M/1 and M/D/1 queueing models to stream processing applications.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

High performance redundant binary multiplier

High performance redundant binary multiplier

引用

2016 international conference on Communication and Signal processing, ICCSP 2016

作者： Bawaskar, Ashish A. Alagdeve, Vilas Keote, Rashmi Electronics Engineering Department Yeshwantrao Chavan College of Engineering Nagpur India

ISBN: (纸本)9781509003969

High speed multiplier designs have been the primacy for multiplier dominated applications such as wireless communications, computer applications, and image processing. In this paper a high performance fixed word length multiplier design by using recently proposed technique to eliminate the error correcting word and a delay efficient parallel prefix Ling adder for final redundant binary to normal binary (RB-NB) conversion has been proposed. These techniques are selected to make achievable tradeoff for area, power and delay. Due to carry-free addition and adaptability, the redundant binary (RB) representation has been picked up in our high-performance multiplier design for partial product summing tree. This multiplier architecture is compared with the design of conventional redundant binary modified booth encoding multiplier (CRBMBE) for area, power and delay analysis. The designed architecture shows improved performance over conventional redundant binary multiplier in terms of area, delay and power-delay product (PDP). © 2016 IEEE.

关键词： Image processing

来源：评论

学校读者我要写书评

暂无评论

Advances in Computing and Information — ICCI '91 1

引用

丛书名： Lecture Notes in Computer Science

1000年

ISBN: (数字)9783540473596

ISBN: (纸本)9783540540298

This volume contains papers presented at the Third international conference on Computing and Information, ICCI '91, held at Carleton University in Ottawa, Canada, May 27-29, 1991. The conference was organized by the School of Computer Science at Carleton University, and was sponsored by the Natural Sciences and Engineering Research Council of Canada (NSERC) and Carleton University. ICCI '91 was an international forum for the presentation of original results in research, development, and applications in computing and information processing. The conference was aimed at both practitioners and theoreticians, and was organized into five streams: - Algorithms and complexity, - Databases and information systems, - parallel processing and systems, - distributed computing and systems, - Expert systems, artificial intelligence. This volume contains three invited papers, by E.C.R. Hehner, R.L. Probert, and S.J. Smith, and 71 selected papers.

关键词： Processor Architectures Artificial Intelligence Models and Principles Information Systems applications (incl. Internet) Algorithm Analysis and Problem Complexity

来源：评论

学校读者我要写书评

暂无评论

Collaborative parallel Local Search for Simplified Protein Structure Prediction

Collaborative Parallel Local Search for Simplified Protein S...

引用

12th IEEE international conference on Trust, Security and Privacy in Computing and Communications (TrustCom)

作者： Rashid, Mahmood A. Newton, M. A. Hakim Hoque, Md Tamjidul Sattar, Abdul Griffith Univ IIIS Nathan Qld 4111 Australia Natl ICT Australia Queensland Res Lab St Lucia Qld Australia Univ New Orleans Dept Comp Sci New Orleans LA 70148 USA

ISBN: (纸本)9780769550220

Protein structure prediction is a challenging optimisation problem to the computer scientists. A large number of existing single-point search algorithms attempt to solve the problem by exploring possible structures and finding the one with the minimum free energy. However, these algorithms perform poorly on large sized proteins due to an astronomically wide search space. In this paper, we present a multi-point local search framework that uses parallel processing techniques to expedite exploration by starting from different points. In our approach, a set of random initial solutions are generated and distributed to different threads. We allow each thread to run for a pre-defined period of time. The improved solutions are stored thread-wise. When the threads finish, the solutions are merged together and the duplicates are removed. A selected distinct set of solutions are then split to different threads again. We tested our approach on large sized proteins for simplified models. The experimental results show that our new parallel framework significantly improves over the results obtained by the state-of-the-art single-point search approaches for the hydrophobicpolar energy model and the three dimensional face-centred-cubic lattice.

关键词： parallel Computing Multi-threading Local Search Protein Structure Prediction HP Model 3D FCC Lattice

来源：评论

学校读者我要写书评

暂无评论

High Throughput Compression of Floating Point Numbers on Graphical processing Units

High Throughput Compression of Floating Point Numbers on Gra...

引用

2nd IEEE international conference on parallel, distributed and Grid Computing (PDGC)

作者： Padyana, Ajith Sudheer, C. D. Baruah, Pallav Kumar Srinivasan, Ashok Sri Sathya Sai Inst Higher Learning Dept Math & Comp Sci Puttaparthi 515134 Andhra Pradesh India Florida State Univ Dept Comp Sci Tallahassee FL 32306 USA

ISBN: (纸本)9781467329255;9781467329224

Compute-intensive tasks in high-end high performance computing (HPC) systems often generate large amounts of data, especially floating-point data, that need to be transmitted over the network. Although computation speeds are very high, the overall performance of these applications is affected by the data transfer overhead. Moreover, as data sets are growing in size rapidly, bandwidth limitations pose a serious bottleneck in several scientific applications. Fast floating point compression can ameliorate the bandwidth limitations. If data is compressed well, then the amount of data transfer is reduced. This reduction in data transfer time comes at the expense of the increased computation required by compression and decompression. It is important for compression and decompression rates to be greater than the network bandwidth;otherwise, it will be faster to transmit uncompressed data directly. Accelerators such as Graphics processing Units (GPU) provide much computational power. In this paper, we show that the computational power of GPUs can be harnessed to provide sufficiently fast compression and decompression for this approach to be effective for data produced by many practical applications. In particularly, we use Holt's Exponential smoothing algorithm from time series analysis, and encode the difference between its predictions and the actual data. This yields a lossless compression scheme. We show that it can be implemented efficiently on GPUs to provide an effective compression scheme for the purpose of saving on data transfer overheads. The primary contribution of this work lies in demonstrating the potential of floating point compression in reducing the 110 bandwidth bottleneck on modern hardware for important classes of scientific applications.

关键词： Throughput Compression Ration Holt's Exponential Smoothening CellBE GPu

来源：评论

学校读者我要写书评

暂无评论

DIFTSAS: a distributed Full Text Search and Analysis System for Big Data

DIFTSAS: a DIstributed Full Text Search and Analysis System ...

引用

IEEE 16th international conference on Computational Science and Engineering (CSE)

作者： Li, Bo Zhang, Jingjie Chen, Mingyu Zhang, JinChao Wang, Kunpeng Meng, Dan Chinese Acad Sci Inst Informat Engn Natl Engn Lab Informat Secur Technol Beijing Peoples R China Beijing Commun Adm Beijing Peoples R China Chinese Acad Sci Comp Applicat Res Ctr Inst Comp Technol Beijing Peoples R China Univ Chinese Acad Sci Beijing Peoples R China

As we enter the big data era, digital world is growing fast and becomes more and more complex in nature, many content-oriented applications require both features of high performance full-text searching and rich functional online data analyzing. In this paper, we exploit the features of both web search and parallel DBMSs to address the challenges to build a scalable, interactive ad-hoc query system for analysis of incrementally accumulated content data, namely DIFTSAS. DIFTSAS is considered as a many-sided system, which is not only capable of handling full-text data as a search engine but also adept at online data analytical processing like a traditional data warehouse. In this paper, we present the design, implementation, and performance evaluation of DIFTSAS.

关键词： full text search engine data warehouse distributed system

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：