检索结果-内蒙古大学图书馆

Annual Conference on neural Information Processing Systems

作者： Daniel Zoran Yair Weiss Interdisciplinary Center for Neural Computation Hebrew University of Jerusalem Israel School of Computer Science and Engineering Hebrew University of Jerusalem Israel

ISBN: (纸本)9781627480031

Simple Gaussian Mixture Models (GMMs) learned from pixels of natural image patches have been recently shown to be surprisingly strong performers in modeling the statistics of natural images. Here we provide an in depth analysis of this simple yet rich model. We show that such a GMM model is able to compete with even the most successful models of natural images in log likelihood scores, denoising performance and sample quality. We provide an analysis of what such a model learns from natural images as a function of number of mixture components — including covariance structure, contrast variation and intricate structures such as textures, boundaries and more. Finally, we show that the salient properties of the GMM learned from natural images can be derived from a simplified Dead Leaves model which explicitly models occlusion, explaining its surprising success relative to other models.

关键词： Gm(m) image patch images natural Death Dead pixel Gauss

来源：评论

学校读者我要写书评

暂无评论

Learning about Canonical Views from Internet Image Collections 12

Learning about Canonical Views from Internet Image Collectio...

引用

Annual Conference on neural Information Processing Systems

作者： Elad Mezuman Yair Weiss Interdisciplinary Center for Neural Computation Edmond & Lily Safra Center for Brain Sciences Hebrew University of Jerusalem School of Computer Science and Engineering Edmond & Lily Safra Center for Brain Sciences Hebrew University of Jerusalem

ISBN: (纸本)9781627480031

Although human object recognition is supposedly robust to viewpoint, much research on human perception indicates that there is a preferred or "canonical" view of objects. This phenomenon was discovered more than 30 years ago but the canonical view of only a small number of categories has been validated experimentally. Moreover, the explanation for why humans prefer the canonical view over other views remains elusive. In this paper we ask: Can we use Internet image collections to learn more about canonical views?We start by manually finding the most common view in the results returned by Internet search engines when queried with the objects used in psychophysical experiments. Our results clearly show that the most likely view in the search engine corresponds to the same view preferred by human subjects in experiments. We also present a simple method to find the most likely view in an image collection and apply it to hundreds of categories. Using the new data we have collected we present strong evidence against the two most prominent formal theories of canonical views and provide novel constraints for new theories.

关键词： image samples Canonical View Search engines Web search engines Learning Human Internets Likely

来源：评论

学校读者我要写书评

暂无评论

On the trade-off between single-neuron complexity and network size with respect to spike-timed computations

引用

BMC Neuroscience 2012年第1期13卷 1-2页

作者： Venkatakrishnan Ramaswamy Arunava Banerjee Interdisciplinary Center for Neural Computation The Edmond and Lily Safra Center for Brain Sciences Hebrew University Jerusalem 91904 Israel Computer and Information Science and Engineering University of Florida Gainesville FL 32611 USA

来源：评论

学校读者我要写书评

暂无评论

NeNMF: An Optimal Gradient Method for Nonnegative Matrix Factorization

引用

IEEE Transactions on Signal Processing 2012年第6期60卷 2882-2898页

作者： Naiyang Guan Dacheng Tao Zhigang Luo Bo Yuan School of Computer Science National University of Defense Technology Changsha China Center for Quantum Computation and Intelligent Systems and the Faculty of Engineering and Information Technology University of Technology Sydney NSW Australia Department of Computer Science and Engineering Shanghai Jiaotong University Shanghai China

Nonnegative matrix factorization (NMF) is a powerful matrix decomposition technique that approximates a nonnegative matrix by the product of two low-rank nonnegative matrix factors. It has been widely applied to signal processing, computer vision, and data mining. Traditional NMF solvers include the multiplicative update rule (MUR), the projected gradient method (PG), the projected nonnegative least squares (PNLS), and the active set method (AS). However, they suffer from one or some of the following three problems: slow convergence rate, numerical instability and nonconvergence. In this paper, we present a new efficient NeNMF solver to simultaneously overcome the aforementioned problems. It applies Nesterov's optimal gradient method to alternatively optimize one factor with another fixed. In particular, at each iteration round, the matrix factor is updated by using the PG method performed on a smartly chosen search point, where the step size is determined by the Lipschitz constant. Since NeNMF does not use the time consuming line search and converges optimally at rate in optimizing each matrix factor, it is superior to MUR and PG in terms of efficiency as well as approximation accuracy. Compared to PNLS and AS that suffer from numerical instability problem in the worst case, NeNMF overcomes this deficiency. In addition, NeNMF can be used to solve -norm, -norm and manifold regularized NMF with the optimal convergence rate. Numerical experiments on both synthetic and real-world datasets show the efficiency of NeNMF for NMF and its variants comparing to representative NMF solvers. Extensive experiments on document clustering suggest the effectiveness of NeNMF.

关键词： Gradient methods Convergence Matrix decomposition Sparse matrices Least squares approximation Educational institutions

来源：评论

学校读者我要写书评

暂无评论

Formal model extraction for combinational equivalence checking

引用

Journal of Convergence Information Technology 2012年第8期7卷 371-380页

作者： Zhang, Guiling Ouyang, Dantong Bai, Hongtao Zeng, Hailin Ma, Tiemin Zhang, Yuehua School of Computer Science and Technology Jilin University Changchun China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Changchun China Center for Computer Fundamental Education Jilin University Changchun China

In formal verification, the first step is to build a formal model. Most of the academic tools represent the designs in BLIF or BLIF-MV network, while cannot directly take Verilog source code as an input. For instance, VIS depends on the translator vl2mv to compile Verilog into the intermediate format BLIF-MV first, and then extracts the FSM (Finite State Models) from it instead of the direct Verilog source code. However, Verilog and VHDL are the most common used hardware description languages for large scale integrated circuits. It is necessary if the source code could be used directly, rather than an intermediate format. We notice that synthesizable Verilog is well-structured which can be represented in SSA (Static Single Assignment) form. In this paper, we describe a method that extracts formal model directly from Verilog source code by converting the code to SSA form. With this method, we have implemented a prototype system that extracts formal model of Verilog programs. We also prove the conversion function-preserving by translating SSA back to Verilog and checking its equivalence with the primary design. To ensure correctness and robustness, we have tested our system with ISCAS and picoJava-II, a specification of microprocessor, and the experimental results on these realistic circuits show that the conversion procedure is function-preserving. Moreover, we take it as the front-end of CEC (Combinational Equivalence Checking) and demonstrate the formal model extracted with our method can be exploited efficiently in CEC.

关键词： computer hardware description languages

来源：评论

学校读者我要写书评

暂无评论

Measuring Steady-State Visual Evoked Potentials from Non-hair-bearing Areas

Measuring Steady-State Visual Evoked Potentials from Non-hai...

引用

Annual International Conference of the IEEE engineering in Medicine and Biology Society

作者： Yu-Te Wang Yijun Wang Chung-Kuan Cheng Tzyy-Ping Jung with Computer Science and Engineering Department University of California San Diego (UCSD) San Diego CA USA with Swartz Center for Computational Neuroscience Institute for Neural Computation and Center for Advanced Neurological Engineering Institute of Engineering in Medicine UCSD San Diego CA USA

ISBN: (纸本)9781424441198

Steady-State Visual Evoked Potential (SSVEP)- based Brain-computer Interface (BCI) applications have been widely applied in laboratories around the world in the recent years. Many studies have shown that the best locations to acquire SSVEPs were from the occipital areas of the scalp. However, for some BCI users such as quadriparetic patients lying face up during ventilation, it is difficult to access the occipital sites. Even for the healthy BCI users, acquiring goodquality EEG signals from the hair-covered occipital sites is inevitably more difficult because it requires skin preparation by a skilled technician and conductive gel usage. Therefore, finding an alternative approach to effectively extract highquality SSVEPs for BCI practice is highly desirable. Since the non-hair-bearing scalp regions are more accessible by all different types of EEG sensors, this study systematically and quantitatively investigated the feasibility of measuring SSVEPs from non-hair-bearing regions, compared to those measured from the occipital areas. Empirical results showed that the signal quality of the SSVEPs from non-hair-bearing areas was comparable with, if not better than, that measured from haircovered occipital areas. These results may significantly improve the practicality of a BCI system in real-life applications;especially used in conjunction with newly available dry EEG sensors.

关键词： Measuring Steady-State Potentials

来源：评论

学校读者我要写书评

暂无评论

Continuous-variable quantum key distribution using thermal states

引用

Physical Review A 2012年第2期86卷 022318-022318页

作者： Christian Weedbrook Stefano Pirandola Timothy C. Ralph Center for Quantum Information and Quantum Control Department of Electrical and Computer Engineering and Department of Physics University of Toronto Toronto M5S 3G4 Canada Department of Computer Science University of York Deramore Lane York YO10 5GH United Kingdom Centre for Quantum Computation and Communication School of Mathematics and Physics University of Queensland St Lucia Queensland 4072 Australia

We consider the security of continuous-variable quantum key distribution using thermal (or noisy) Gaussian resource states. Specifically, we analyze this against collective Gaussian attacks using direct and reverse reconciliation where both protocols use either homodyne or heterodyne detection. We show that in the case of direct reconciliation with heterodyne detection, an improved robustness to channel noise is achieved when large amounts of preparation noise is added, as compared to the case when no preparation noise is added. We also consider the theoretical limit of infinite preparation noise and show a secure key can still be achieved in this limit provided the channel noise is less than the preparation noise. Finally, we consider the security of quantum key distribution at various electromagnetic wavelengths and derive an upper bound related to an entanglement-breaking eavesdropping attack and discuss the feasibility of microwave quantum key distribution.

关键词： CONTINUOUS functions VARIABLES (Mathematics) QUANTUM theory ENERGY levels (Quantum mechanics) GAUSSIAN distribution ROBUST control

来源：评论

学校读者我要写书评

暂无评论

High-order local dynamic programming

High-order local dynamic programming

引用

作者： Tassa, Yuval Todorov, Emanuel Interdisciplinary Center for Neural Computation Hebrew University Jerusalem Israel Applied Mathematics and Computer Science and Engineering University of Washington Seattle United States

ISBN: (纸本)9781424498888

We describe a new local dynamic programming algorithm for solving stochastic continuous Optimal Control problems. We use cubature integration to both propagate the state distribution and perform the Bellman backup. The algorithm can approximate the local policy and cost-to-go with arbitrary function bases. We compare the classic quadratic cost-to-go/linear-feedback controller to a cubic cost-to-go/quadratic policy controller on a 10-dimensional simulated swimming robot, and find that the higher order approximation yields a more general policy with a larger basin of attraction. © 2011 IEEE.

关键词： Dynamic programming

来源：评论

学校读者我要写书评

暂无评论

Mining item-item and between-set correlated association rules

引用

Journal of Zhejiang University-science C(computers and Electronics) 2011年第2期12卷 96-109页

作者： Bin SHEN Min YAO Li-jun XIE Rong ZHU Yun-ting TANG Ningbo Institute of Technology Zhejiang University Ningbo 315100 China School of Computer Science and Technology Zhejiang University Hangzhou 310027 China Center for Engineering & Scientific Computation School of Aeronautics and Astronautics Zhejiang University Hangzhou 310027 China

To overcome the failure in eliminating suspicious patterns or association rules existing in traditional association rules mining, we propose a novel method to mine item-item and between-set correlated association rules. First, we present three measurements: the association, correlation, and item-set correlation measurements. In the association measurement, the all-confidence measure is used to filter suspicious cross-support patterns, while the all-item-confidence measure is applied in the correlation measurement to eliminate spurious association rules that contain negatively correlated items. Then, we define the item-set correlation measurement and show its corresponding properties. By using this measurement, spurious association rules in which the antecedent and consequent item-sets are negatively correlated can be eliminated. Finally, we propose item-item and between-set correlated association rules and two mining algorithms, I&ISCoMine_AP and I&ISCoMine_CT. Experimental results with synthetic and real retail datasets show that the proposed method is effective and valid.

关键词： Item-item and between-set correlated association rules All-confidence All-item-confidence Item-set correlation Mining algorithms Pruning effect

来源：评论

学校读者我要写书评

暂无评论

Online nonnegative matrix factorization with robust stochastic approximation

引用

IEEE transactions on neural networks and learning systems 2012年第7期23卷 1087-99页

作者： Naiyang Guan Dacheng Tao Zhigang Luo Bo Yuan School of Computer Science National University of Defense Technology Changsha China Center for Quantum Computation & Intelligent Systems University of Technology Sydney NSW Australia Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai China

Nonnegative matrix factorization (NMF) has become a popular dimension-reduction method and has been widely applied to image processing and pattern recognition problems. However, conventional NMF learning methods require the entire dataset to reside in the memory and thus cannot be applied to large-scale or streaming datasets. In this paper, we propose an efficient online RSA-NMF algorithm (OR-NMF) that learns NMF in an incremental fashion and thus solves this problem. In particular, OR-NMF receives one sample or a chunk of samples per step and updates the bases via robust stochastic approximation. Benefitting from the smartly chosen learning rate and averaging technique, OR-NMF converges at the rate of in each update of the bases. Furthermore, we prove that OR-NMF almost surely converges to a local optimal solution by using the quasi-martingale. By using a buffering strategy, we keep both the time and space complexities of one step of the OR-NMF constant and make OR-NMF suitable for large-scale or streaming datasets. Preliminary experimental results on real-world datasets show that OR-NMF outperforms the existing online NMF (ONMF) algorithms in terms of efficiency. Experimental results of face recognition and image annotation on public datasets confirm the effectiveness of OR-NMF compared with the existing ONMF algorithms.

关键词： Complexity theory Convergence Approximation methods Algorithm design and analysis Robustness Approximation algorithms Streaming media

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：