检索结果-内蒙古大学图书馆

Generation of a compact code for extended binary memoryless sources with low entropy

ELECTRONICS AND COMMUNICATIONS IN JAPAN PART III-FUNDAMENTAL ELECTRONIC SCIENCE 2000年第1期83卷 70-77页

作者： Hasegawa, M Kato, S Yamada, Y Utsunomiya Univ Fac Engn Utsunomiya Tochigi 321 Japan

A method is proposed for configuring the compact code for extended binary memoryless sources with probabilistic deviations in symbol appearance. In an n-th order extended information source for binary memoryless sources, messages with identical appearance probabilities are generated abundantly. If these messages are grouped for treatment, no rearrangement of the messages encompassing the groups is generated in degenerate operations by the Huffman algorithm for low-entropy information sources where the appearance probability of the superior symbols approaches arbitrarily close to 1. That is the situation is used in this paper. It is reported that the code length and sets of code words, average code length, and maximum code length of the compact code can be derived by a simple equation without forming a code tree. (C) 1999 Scripta Technica.

关键词： compact code Huffman code information source coding memoryless information source expanded information source

来源：评论

学校读者我要写书评

暂无评论

Generation of a compact code for extended binary memoryless sources with low entropy

引用

Electronics and Communications in Japan (Part III: Fundamental Electronic Science) 1999年第1期83卷

作者： Madoka Hasegawa Shigeo Kato Yoshifumi Yamada Faculty of Engineering Utsunomiya University Utsunomiya Japan 321

A method is proposed for configuring the compact code for extended binary memoryless sources with probabilistic deviations in symbol appearance. In an n -th order extended information source for binary memoryless sources, messages with identical appearance probabilities are generated abundantly. If these messages are grouped for treatment, no rearrangement of the messages encompassing the groups is generated in degenerate operations by the Huffman algorithm for low-entropy information sources where the appearance probability of the superior symbols approaches arbitrarily close to 1. That is the situation used in this paper. It is reported that the code length and sets of code words, average code length, and maximum code length of the compact code can be derived by a simple equation without forming a code tree. © 1999 Scripta Technica, Electron Comm Jpn Pt 3, 83(1): 70–77, 2000

关键词： compact code Huffman code information source coding memoryless information source expanded information source

来源：评论

学校读者我要写书评

暂无评论

The Supertree of the compact codes

The Supertree of the Compact Codes

引用

International Symposium on Telecommunications

作者： Narimani, H. Khosravifard, M. Isfahan Univ Technol Dept Elect & Comp Engn Esfahan *** Iran

ISBN: (纸本)9781424427505

An instantaneous D-ary code which minimizes the average codeword length for an information source is called a compact code. It is known that for a D-ary compact code with codeword lengths l(1),l(2), ..., l(n) (where n is the form of n = k (D - 1) + 1 for some positive integer k), we have Sigma D(-l1) = 1. Since construction of n D-ary codewords given codeword lengths l(1),l(2), ... ,l(n) is a straight forward task, we generate all possible codeword length sequences. In this paper, we present an algorithm which gets all compact codes with k(D-1)+1 codewords and generates all compact codes with (k+1)(D-1)+1 codewords. Based on this algorithm, ST(n)(D), the Supertree or all D-ary compact codes, is introduced. Any node in the m-th level of ST(n)(D) is associated with a unique compact code with 2D-1+m(D-1) codewords. Following the proposed approach, any D-ary compact code with n codewords can be represented by [(n - 1)/(D - 1)] - 2 bits.

关键词： Huffman algorithm compact code Kraft inequality multiplicity sequence Supertree

来源：评论

学校读者我要写书评

暂无评论

Exploitation of image statistics with sparse coding in the case of stereo vision

引用

NEURAL NETWORKS 2021年 135卷 158-176页

作者： Ecke, Gerrit A. Papp, Harald M. Mallot, Hanspeter A. Univ Tubingen Dept Biol Cognit Neurosci Unit Morgenstelle 28 D-72076 Tubingen Germany Mercedes Benz AG Leibnizstr 2 D-71032 Boblingen Germany

The sparse coding algorithm has served as a model for early processing in mammalian vision. It has been assumed that the brain uses sparse coding to exploit statistical properties of the sensory stream. We hypothesize that sparse coding discovers patterns from the data set, which can be used to estimate a set of stimulus parameters by simple readout. In this study, we chose a model of stereo vision to test our hypothesis. We used the Locally Competitive Algorithm (LCA), followed by a naive Bayes classifier, to infer stereo disparity. From the results we report three observations. First, disparity inference was successful with this naturalistic processing pipeline. Second, an expanded, highly redundant representation is required to robustly identify the input patterns. Third, the inference error can be predicted from the number of active coefficients in the LCA representation. We conclude that sparse coding can generate a suitable general representation for subsequent inference tasks. (c) 2020 Published by Elsevier Ltd.

关键词： Sparse coding Locally Competitive Algorithm (LCA) Efficient coding compact code Probabilistic inference Stereo vision

来源：评论

学校读者我要写书评

暂无评论

Shared Predictive Cross-Modal Deep Quantization

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2018年第11期29卷 5292-5303页

作者： Yang, Erkun Deng, Cheng Li, Chao Liu, Wei Li, Jie Tao, Dacheng Xidian Univ Sch Elect Engn Xian 710071 Shaanxi Peoples R China Tencent AI Lab Shenzhen 518057 Peoples R China Univ Sydney UBTECH Sydney Artificial Intelligence Ctr Fac Engn & Informat Technol Darlington NSW 2008 Australia Univ Sydney Sch Informat Technol Fac Engn & Informat Technol Darlington NSW 2008 Australia

With explosive growth of data volume and ever-increasing diversity of data modalities, cross-modal similarity search, which conducts nearest neighbor search across different modalities, has been attracting increasing interest. This paper presents a deep compact code learning solution for efficient cross-modal similarity search. Many recent studies have proven that quantization-based approaches perform generally better than hashing-based approaches on single-modal similarity search. In this paper, we propose a deep quantization approach, which is among the early attempts of leveraging deep neural networks into quantization-based cross-modal similarity search. Our approach, dubbed shared predictive deep quantization (SPDQ), explicitly formulates a shared subspace across different modalities and two private subspaces for individual modalities, and representations in the shared subspace and the private subspaces are learned simultaneously by embedding them to a reproducing kernel Hilbert space, where the mean embedding of different modality distributions can be explicitly compared. In addition, in the shared subspace, a quantizer is learned to produce the semantics preserving compact codes with the help of label alignment. Thanks to this novel network architecture in cooperation with supervised quantization training, SPDQ can preserve intramodal and intermodal similarities as much as possible and greatly reduce quantization error. Experiments on two popular benchmarks corroborate that our approach outperforms state-of-the-art methods.

关键词： compact code deep learning multimodal private network quantization shared network

来源：评论

学校读者我要写书评

暂无评论

Distance Encoded Product Quantization for Approximate K-Nearest Neighbor Search in High-Dimensional Space

引用

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 2019年第9期41卷 2084-2097页

作者： Heo, Jae-Pil Lin, Zhe Yoon, Sung-Eui Sungkyunkwan Univ Suwon 16419 South Korea Adobe Res San Jose CA 95110 USA Korea Adv Inst Sci & Technol Daejeon 34141 South Korea

Approximate K-nearest neighbor search is a fundamental problem in computer science. The problem is especially important for high-dimensional and large-scale data. Recently, many techniques encoding high-dimensional data to compact codes have been proposed. The product quantization and its variations that encode the cluster index in each subspace have been shown to provide impressive accuracy. In this paper, we explore a simple question: is it best to use all the bit-budget for encoding a cluster index? We have found that as data points are located farther away from the cluster centers, the error of estimated distance becomes larger. To address this issue, we propose a novel compact code representation that encodes both the cluster index and quantized distance between a point and its cluster center in each subspace by distributing the bit-budget. We also propose two distance estimators tailored to our representation. We further extend our method to encode global residual distances in the original space. We have evaluated our proposed methods on benchmarks consisting of GIST, VLAD, and CNN features. Our extensive experiments show that the proposed methods significantly and consistently improve the search accuracy over other tested techniques. This result is achieved mainly because our methods accurately estimate distances.

关键词： Vector quantization nearest neighbor search image retrieval compact code high-dimensional search

来源：评论

学校读者我要写书评

暂无评论

WHAT IS THE GOAL OF SENSORY CODING

引用

NEURAL COMPUTATION 1994年第4期6卷 559-601页

作者： FIELD, DJ Department of Psychology Cornell University Ithaca NY 14850 USA

A number of recent attempts have been made to describe early sensory coding in terms of a general information processing strategy. In this paper, two strategies are contrasted. Both strategies take advantage of the redundancy in the environment to produce more effective representations. The first is described as a ''compact'' coding scheme. compact code performs a transform that allows the input to be represented with a reduced number of vectors (cells) with minimal RMS error. This approach has recently become popular in the neural network literature and is related to a process called Principal Components Analysis (PCA). A number of recent papers have suggested that the optimal ''compact'' code for representing natural scenes will have units with receptive field profiles much like those found in the retina and primary visual cortex. However, in this paper, it is proposed that compact coding schemes are insufficient to account for the receptive field properties of cells in the mammalian visual pathway. In contrast, it is proposed that the visual system is near to optimal in representing natural scenes only if optimality is defined in terms of ''sparse distributed'' coding. In a sparse distributed code, all cells in the code have an equal response probability across the class of images but have a low response probability for any single image. In such a code, the dimensionality is not reduced. Rather, the redundancy of the input is transformed into the redundancy of the firing pattern of cells. It is proposed that the signature for a sparse code is found in the fourth moment of the response distribution (i.e., the kurtosis). In measurements with 55 calibrated natural scenes, the kurtosis was found to peak when the bandwidths of the visual code matched those of cells in the mammalian visual cortex. codes resembling ''wavelet transforms'' are proposed to be effective because the response histograms of such codes are sparse (i.e., show high kurtosis) when presented with n

关键词： .visual cortex Goal of Sensory Ted Adeison PsychoIogt Cornell Uniz compact code representing natural scenes

来源：评论

学校读者我要写书评

暂无评论

Thermodynamic-like transformations in information theory

引用

Journal of Statistical Physics 1969年第1期1卷 107-131页

作者： Reiss, H. Department of Chemistry University of California Los Angeles United States

This article explores the application of thermodynamic and statistical thermodynamic formalism to information theory problems. In particular, the applicability of the transformation theory of thermodynamics is investigated. After a brief tutorial discussion of thermodynamic and statistical thermodynamic methods and concepts, their information theory analogues are developed. Besides information theory entropy, the information theory counterparts of temperature, chemical potential, Helmholtz free energy, etc., are developed and related to conventional information theory concepts such as channel capacity, matching of source and channel, etc. Information theory theorems are proved via the statistical thermodynamic analogue method;and, finally, several problems are formulated and solved using thermodynamic-like transformations. This article is aimed chiefly at bridging the interface between the two disciplines, and is intended to be provocative. Therefore, no attempt has been made to have it be all inclusive. © 1969 Plenum Publishing Corporation.

关键词： Channel capacity compact code Information theory Matching Noiseless channel Partition function Statistical thermodynamics Thermodynamic transformations

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：