Search

SVM is a new general learning method based on the statistic learning system which can be used as an effective means to process small sample, nonlinear and high dimensional pattern recognition. This paper did research on the learning algorithm of support vector machine, extracted characteristic data of banknote which is on account of PCA according to the characteristics of the support vector machine (SVM), and proposed to put support vector machine (SVM) into banknotes denomination recognition by combining SMO training algorithm with one-to-many multi-value classification algorithm. Besides, this article used genetic algorithm in parameters optimization such as the punishment coefficient C of soft margin SVM and the width parameter of Gaussian kernel function. The ultimate purpose is to recognize the denomination of banknote efficiently and accurately. The experimental results verified that this kind of recognition method increases the recognition accuracy up to 90% or more.

3D geo-information can be thought of in several ways. At the simplest level it involves a 2D data structure with elevation attributes, as with remote sensing data such as LIDAR. The resulting structure forms a simple 2-manifold. At a slightly more advanced level we may recognise that the earth may not always be modelled by a planar graph, but requires bridges and tunnels. This 2-manifold of higher genus may still use the same data structure (e.g. a triangulation) but certain assumptions (e.g. a Delaunay triangulation) no longer hold. Finally, we may wish to model true volumes, in which case a triangulation might be replaced by a tetrahedralisation.

We present an algorithm to find a flat folding of a piece of paper, so that one complete straight cut on the folding creates any desired plane graph of cuts. The folds are based on the straight skeleton, which lines up the desired edges by folding along various bisectors; and a collection of perpendiculars that make the crease pattern foldable. We prove that the crease pattern is flat foldable by demonstrating a family of folded states with the desired properties.

We report an experience on a practical system for drawing hypergraphs in the subset standard. The PATATE system is based on the application of a classical force directed method to a dynamic graph, which is deduced, at a given iteration time, from the hypergraph structure and particular vertex locations. Different strategies to define the dynamic underlying graph are presented. We illustrate in particular the method when the graph is obtained by computing an Euclidean Steiner tree.

We describe an architecture requiring very few changes to any standard von Neumann machine that provably withstands coalitions between a malicious operating system and other users, in the sense that: If the operating system permits a program to run, then the program produces the same outputs as it would produce if it were running on an ideal, single-user machine; moreover, even if the operating system behaves according to expectations only most of the time, the programs get executed. The only information leaked by a program to the malicious coalition is the time and space requirements of the program. If the malicious operating system is dynamically replaced by a good operating system, then the latter can quickly and correctly determine what memory resources are available for future programs, as well as how much time is left for each of the currently executing programs, and can distribute these resources without any restrictions. This can be accomplished without restarting currently executing programs. To our knowledge, ours is the first attempt to provide provable guarantees along these lines, and the first treatment of any kind, provable or otherwise, for the third property.

This paper addresses the problem of shape from shadings under perspective projection and turntable motion.Two-Frame-Theory is a newly proposed method for 3D shape recovery. It estimates shape by solving a first order quasi-linear partial differential equation through the method of characteristics. One major drawback of this method is that it assumes an orthographic camera which limits its application. This paper re-examines the basic idea of the Two-Frame-Theory under the assumption of a perspective camera, and derives a first order quasi-linear partial differential equation for shape recovery under turntable motion. The Dirichlet boundary condition is derived based on Dynamic programming. The proposed method is tested against synthetic and real data. Experimental results show that perspective projection can be used in the framework of Two-Frame-Theory, and competitive results can be achieved.

A natural way to draw two planar graphs whose vertex sets are matched is to assign each matched pair a unique y-coordinate. In this paper we introduce the concept of such matched drawings, which are a relaxation of simultaneous geometric embeddings with mapping. We study which classes of graphs allow matched drawings and show that (i) two 3-connected planar graphs or a 3-connected planar graph and a tree may not be matched drawable, while (ii) two trees or a planar graph and a planar graph of some special families—such as unlabeled level planar (ULP) graphs or the family of "carousel graphs"—are always matched drawable. Research partially supported by the MIUR Project "MAINSTREAM: Algorithms for massive information structures and data streams".

In order to inherit and develop the Couplet culture, one of the most famous language culture in China, this paper implements a creation system of couplets. In this paper, a noval way to generate the customized couplet is discussed from the name-embedded couplet. The 2-gram word graph technology is used to generate the first sentence which contained the given word. Then, combined with language model which based on statistical method, grammar model and mutual information model, improved Viterbi Algorithm based on HMM is used to decode the second sentence. It turned out that the system can realize the automatic generation of the name embedded couplet. According to the contents involved in the given couplet, the system can generate pictures to match the couplet. This system is aimed for Spring Festival couplets and Birthday couplets, both are name-embedded couplet that generated with the name of people. Users can choose the templates and relative special elements freely to create a satisfactory multimedia couplet picture.

Electronic payment systems have leveraged the advantages offered by the RFID technology, whose security is promised to be improved by applying the notion of Physically Unclonable Functions (PUFs). Along with the evolution of PUFs, numerous successful attacks against PUFs have been proposed in the literature. Among these are machine learning (ML) attacks, ranging from heuristic approaches to provable algorithms, that have attracted great attention. Our paper pursues this line of research by introducing a Fourier analysis based attack against PUFs. More specifically, this paper focuses on two main aspects of ML attacks, namely being provable and noise tolerant. In this regard, we prove that our attack is naturally integrated into a provable Probably Approximately Correct (PAC) model. Moreover, we show that our attacks against known PUF families are effective and applicable even in the presence of noise. Our proof relies heavily on the intrinsic properties of these PUF families, namely arbiter, Ring Oscillator (RO), and Bistable Ring (BR) PUF families. We believe that our new style of ML algorithms, which take advantage of the Fourier analysis principle, can offer better measures of PUF security.

We briefly present notions and results from three directions of research which use formal language theory tools for modelling operations specific to DNA (and RNA) recombinations; in all cases one obtains computability models which are universal (language generating devices are obtained which are equivalent in power with Turing machines). The basic operations are those of sticking (a model of the Watson-Crick complementarity), of splicing (a model of the recombinant behaviour of DNA sequences under the influence of restriction enzymes), and of insertion/deletion (known to hold both for DNA and for RNA sequences). Because this paper is an overview, we present here only definitions and results without proofs; further details can be found in the papers quoted below.

Based on the measured data of IGS, the characteristics of QZSS signal, namely, carrier to noise ratio, noise and multipath characteristics, inter-frequency bias, were studied in this paper. At the same time, the results were compared with those of other satellite navigation systems, such as GPS and Galileo, to analyze the reason for the differences. Methods and results provided in this paper have important reference values for the study and construction of satellite navigation systems.

This paper describes a new method for error model construction. Instead of the standard local slope analysis of the Allan variance, two major modifications are proposed: (1) the Direct Bound principle, i.e. finding an entire error model that generates analyzing tool values that tightly bound the analyzing tool values generated by the real data; (2) instead of using Allan variance as a unique analyzing tool, a variety of analyzing tools termed Direct Predictor (DP) types 0, 1, 2, and 3 are introduced. In the paper, a uniform structure of DPs is developed and their parameterization is extended. For a nominal model that consists of a Markov process with additive white noise, the analytical functions for DPs are presented (for infinite data length). The errors due to the final data length are analyzed. Using these results, a reliable optimization problem is presented to implement the Direct Bound approach. The flexibility of working with hard and soft bounds is introduced. The presented simulation results show that the proposed method is indeed efficient and provides satisfactory results for model parameter estimations. The paper concludes with a description of an entire engineering process to cover test design and its analysis.

Large time-delay often occurs unexpectedly in data transmission via Internet. As a kind of typical non-Gaussian stochastic process cannot be described by traditional prediction models with Gaussian innovation. In this paper, FARIMA model with standard symmetrical α stable distribution is adopted to model the transmission time-delay in Internet, and the method for estimating the model parameters is presented. Numerical results show that the approach proposed in this paper is more accurate than the general FARIMA model with Gaussian innovation in modeling Internet time-delay.

One of the most popular models widely used in nanotribology as the basis for many investigations of frictional mechanisms on the atomic scale is the so-called Tomlinson model, consisting of a point mass driven over a periodic potential. The name “Tomlinson model” is, however, historically incorrect: The paper by Tomlinson from the year 1929 which is often cited in this context did not, in fact, contain the model known as the “Tomlinson model” and suggests, instead, an adhesive contribution to friction. In reality, it was Ludwig Prandtl who suggested this model in 1928 to describe the plastic deformation in crystals and dry friction. Staying in line with some other researchers, we call this model the “Prandtl-Tomlinson model,” although the model could simply and rightly be dubbed the “Prandtl model.” The original paper by Ludwig Prandtl was written in German and was not accessible for a long time for the largest part of the international tribological community. The present paper is a historical introduction to the English translation of the classical paper by Ludwig Prandtl, which was published in 2012. It gives a short review of the model as well as its properties, applications, and extensions from a contemporary point of view.

This paper investigated the use of instantaneous frequency (IF) instead of power amplitude and power spectrum in side-channel analysis. By opposition to the constant frequency used in Fourier Transform, instantaneous frequency reflects local phase differences and allows detecting frequency variations. These variations reflect the processed binary data and are hence cryptanalytically useful. IF exploits the fact that after higher power drops more time is required to restore power back to its nominal value. Whilst our experiments reveal IF does not bring specific benefits over usual power attacks when applied to unprotected designs, IF allows to obtain much better results in the presence of amplitude modification countermeasures.

This paper proposes trajectory tracking algorithm for differential drive type of Automatic Guided Vehicle (AGV) system with the unknown wheel radii using adaptive backstepping control method. To guarantee the tracking errors go to zero, backstepping control method is proposed. By choosing appropriate Lyapunov function based on its kinematic modeling, system stability is guaranteed and a control law can be obtained. In this paper, the unknown radii of left and right wheels caused by uneven load distribution or manufacturing imperfection are considered. To solve this problem, an adaptive law is proposed to estimate the changing of wheels radii. The simulation and experimental results show that the proposed controller successfully estimates the unknown parameters and tracks the reference trajectories.

Not only the world is distributed, but more and more applications are distributed. Hence, a fundamental question is the following one: What can be computed in a distributed system? The answer to this question depends on the environment in which evolves the considered distributed system, i.e., on the assumptions the system relies on. This environment is very often left implicit and nearly always not formulated in terms of precise underlying requirements. In the extreme case where the environment is such that there is no synchrony assumption and the computing entities may commit failures, many problems become impossible to solve (in these cases, a network of Turing machines where some machines may crash, is less powerful than a single reliable Turing machine). Given a distributed computing problem, it is consequently important to know the weakest assumptions (lower bounds) that give the limits beyond which the considered distributed problem cannot be solved. This paper is a short introduction to this kind of issues. It first presents a few of elements related to distributed computability, and then briefly addresses distributed complexity issues. The style of the paper is voluntarily informal.

Since human face is often not clear in the surveillance video, this paper proposes a retrieval method on the whole body with fine-grained feature extraction. This method first extracts foreground region of human movement based on Gaussian mixed model (GMM). The human body is divided into two parts, head and below the head, based on the human morphological features and skin color. There are three parts coupled with the whole body, and then, we extract color feature for each part. Secondly, the human body is divided into front and back feature samples according to the a priori knowledge. Calculate the gap between the retrieval eigenvectors and the eigenvectors of the target body, then determine whether match. The experimental results show that this method maintains better retrieval precision when the recall rate is high. This paper target retrieve is real time on video.

Signal de-noising and extraction of useful signal feature are significant problems in the work of scientific research. Higher-order cumulants (HOC) have a strong ability of noise reduction. However, the increase in the order number of HOC will lead to the increase in computation. This will bring great difficulties to practical application. This paper introduces the knowledge of 1.5-dimensional spectrum, which results from the HOC. 1.5-dimensional spectrum is actually a simplified calculation method of HOC. It also remains the HOC’s excellent characteristic which can suppress the additive Gaussian noise. Consequently, 1.5-dimensional spectrum can be well applied in engineering practice. Meanwhile, the Hilbert transform is also introduced simply in this paper. The simulation signals and rolling failure data are processed with the 1.5-dimensional spectrum and Hilbert transform, respectively. The comparison results confirm the practical value of 1.5-dimensional spectrum.

This paper examines the impulse control of a standard Brownian motion under a long-term average criterion. In contrast with the dynamic programming approach, this paper first imbeds the stochastic control problem into an infinite-dimensional linear program over a space of measures and then reduces the problem to a simpler nonlinear optimization that has a familiar interpretation. One is able to easily identify the optimal cost and a family of optimal impulse control policies.

CAPTCHA is a completely automated public Turing test which has been broadly adopted to enhance the website security and strengthen its attack resistance ability. The research of recognition technology of CAPTCHA has a great benefit for increasing the machine learning ability and aiding to designing high security code. The work in this paper focuses on the most difficult adhesion code in picture. At present, the traditional methods which depend heavily on segmentation will become invalid for most adhesion picture code. To resolve the problem, an integrated recognition algorithm based on mask matching is presented in this paper. The recognition of picture code is completed regardless of segmentation and noise removal beforehand. Especially, the matching degree is controlled by a fuzzy factor which can be adjusted automatically to meet with the picture code. During mask creation, two kinds of methods are provided. Finally, a series of experiments from different websites demonstrate that the algorithm can crack all kinds of picture codes and has a satisfied performance in time and popularity.

In this paper, we model the multilevel statistical structure as Hierarchical Hidden Markov Models (HHMM) for the problem of predicting the state of human behavior based on user profile in a ubiquitous home network. Algorithms to analyze the behavioral patterns of a user using the information provided by the user in a home network system. We propose the detecting of abnormal behavior algorithm, which builds profile based on the actions taken when the user enters a room. The main contributions of this paper lie in the application of the shared structure HHMM, the estimation of the state of a user’s behavior, and the detection of abnormal behavior. The user behavior data from an experiment show that directly modeling shared structures improves the recognition efficiency and prediction accuracy for the state of a human’s behavior when compared with a flat HMM.

Over the past few years, microblogging websites, such as Twitter, are growing increasingly popular. Different with traditional medias, tweets are structured data and with a lot of noisy words. Topic modeling algorithms for traditional medias have been studied well, but our understanding of Twitter still remains limited and few algorithms are specially designed to mine Twitter data according to its own characteristics. Previous studies usually employ only one type of topic to analyze hot topics of the Twitter community and are greatly affected by the large amount of noisy words in tweets. We have observed that, in the Twitter community, users tend to discuss two types of topics actually. One mainly focuses on their personal lives and the other on hot issues of the society. These two types of topics usually yield different distributions. In this paper, we introduce the Categorical Topic Model. This model incorporates the features of Twitter data to divide topics into two types in semantic and introduce a word distribution for background words to filter out noisy words. Our model is able to discover different types of topics efficiently, indicate which topics are interested by an user and find hot issues of the Twitter community. Employing the Gibbs sampling, we compare our model with Latent Dirichlet Allocation and Author Topic Model on the TREC2011 data set and examples of discovered public topics and personal topics are also discussed in our paper.

In this paper, we focus on the issue of characterizing microblogs' topics based on topic models. Different from dealing with traditional textual media (such as news documents), modeling microblogs has three challenges: 1) too much noise; 2) short text; and 3) content incompleteness. Previously, all these limitations have been investigated separately. Some work filters the noise through a prior classification; some enhances the text through the user's blog history; and some utilizes the social network. However, none of these work could solve all the above limitations simultaneously. To solve this problem, we make a combination of previous work in this paper, and propose a unified generative model for characterizing microblogs' topics. In the proposed unified approach, all the three limitations could be solved. A collapsed Gibbs-sampling optimization method is derived for estimating the parameters. Through both qualitative and quantitative analysis in Twitter, we demonstrate that our approach consistently outperforms previous methods at a significant scale.

In this paper, a new method for reducing noise within chaotic signal based on ICA (Independent Component Analysis) and EMD (Empirical Mode Decomposition) is proposed. The basic idea is decomposing chaotic signal and constructing multidimensional input vectors, firstly, on the base of EMD and its translation invariance. Then, it makes the independent component analysis on the input vectors, which means a self adapting denoising is carried out for the intrinsic mode functions (IMFs) of chaotic signal. At last, all the IMFs composed the new denoised chaotic signal. An experiment on Lorenz chaotic signal which is composed as different Gaussian noises and monthly observed chaotic sequence on sunspots was put into practice. The result proved the method proposed by this paper is effective in denoising of chaotic signal. Moreover, it can correct the center point in the phase space effectively, which makes it approach the real track of the chaotic attractor.

A multilayer perceptron comprising a single hidden layer of neurons with sigmoidal transfer functions can approximate any computable function to arbitrary accuracy. The size of the hidden layer dictates the approximation capability of the multilayer perceptron and automatically determining a suitable network size for a given data set is an interesting question. This paper considers the problem of inferring the size of multilayer perceptron networks with the MMLD model selection criterion which is based on the minimum message length principle. The two main contributions of the paper are: (1) a new model selection criterion for inference of fully-connected multilayer perceptrons in regression problems, and (2) an efficient algorithm for computing MMLD-type codelengths in mathematically challenging model classes. Empirical performance of the new algorithm is demonstrated on artificially generated and real data sets.

This paper mainly researches on how members of dynamic alliance make a cooperative game under manufacturing grid environment. Determined the contribution level which is made by resources suppliers to demanders by adjusting the proportion of resources provided by suppliers, so that to maximize the total revenue of this alliance. First, this paper establishes a differential dynamic equation for members of this alliance. Then it determines the performance index function of alliance’s total revenue. Finally, by using the pontryagin’s maximum principle to solve this problem, it gets the Nash equilibrium and through an example to verify the feasibility of this method.

The purpose of this paper is to introduce and study Ishikawa iterative algorithms for solving the SFP in the setting of infinite-dimensional Hilbert spaces. The main results presented in this paper improve and extend some recent results done by Xu [Iterative methods for the split feasibility problem in infinite-dimensional Hilbert space, Inverse Problems 26 (2010) 105018]. At the end we prove that the accumulation of errors in Ishikawa iterative CQ algorithm is bounded in certain range.

A central computational problem for analyzing and model checking various classes of infinite-state recursive probabilistic systems (including quasi-birth-death processes, multi-type branching processes, stochastic context-free grammars, probabilistic pushdown automata and recursive Markov chains) is the computation of termination probabilities, and computing these probabilities in turn boils down to computing the least fixed point (LFP) solution of a corresponding monotone polynomial system (MPS) of equations, denoted x=P(x). It was shown by Etessami and Yannakakis [11] that a decomposed variant of Newton's method converges monotonically to the LFP solution for any MPS that has a non-negative solution. Subsequently, Esparza, Kiefer, and Luttenberger [7] obtained upper bounds on the convergence rate of Newton's method for certain classes of MPSs. More recently, better upper bounds have been obtained for special classes of MPSs ([10, 9]). However, prior to this paper, for arbitrary (not necessarily strongly-connected) MPSs, no upper bounds at all were known on the convergence rate of Newton's method as a function of the encoding size |P| of the input MPS, x=P(x). In this paper we provide worst-case upper bounds, as a function of both the input encoding size |P|, and e>0, on the number of iterations required for decomposed Newton's method (even with rounding) to converge to within additive error e>0 of q*, for an arbitrary MPS with LFP solution q*. Our upper bounds are essentially optimal in terms of several important parameters of the problem. Using our upper bounds, and building on prior work, we obtain the first P-time algorithm (in the standard Turing model of computation) for quantitative model checking, to within arbitrary desired precision, of discrete-time QBDs and (equivalently) probabilistic 1-counter automata, with respect to any (fixed) ω-regular or LTL property.

The paper addresses development of a brain-like system based on Lamarckian view toward evolution. It describes a development of an artificial brain, from an artificial genome, through a neural stem cell. In the presented design a modulon level of genetic hierarchical control is used. In order to evolve such a system, two environments are considered, genetic and behavioral. The genome comes from the genetic environment, evolves into an artificial brain, and then updates the memory through interaction with the behavioral environment. The updated genome can then be sent back to the genetic environment. The memory units of the artificial brain are synaptic weights which in this paper represent achievement motivations of the agent, so they are updated in relation to particular achievement. A simulation of the process of learning by updating achievement motivations in the behavioral environment is also shown.

This paper focuses on performance improvements of the permanent magnet synchronous motor (PMSM) vector control. In this paper, a novel second order sliding mode control (SOSMC) algorithm is presented to accomplish velocity control of the PMSM. The integral manifold is utilized to avoid noise signals being amplified because of the acceleration information in control system, and the second order sliding mode control law is achieved by a Lyapunov function approach. The novel method can successfully eliminate the system chattering problem and improve the performance of the PMSM control system, such as fast response, high robustness and tracking speed performance. Meanwhile, an anti-windup control method is used to solve the problem of the windup phenomenon of the PMSM control system. The experimental results show that the proposed method is feasible and effective and is capable of controlling the permanent magnet synchronous motor.

Given two pictures of the same scene captured using the same camera and the same lens, the one captured with a large aperture will appear partially blurred while the other captured with a small aperture will appear totally sharp. This paper investigates two possible ways of inferring depth of the scene from such an image pair with the constraint that both pictures are focused on the closest point of the scene. Our first method uses a series of Gaussian kernels to blur the image pair, and in the second method, the image pair will be shrunk to a series of smaller dimensions. In both methods, sharp areas in both images will always stay similar to each other, whereas the areas that appear sharp in one image but blurred in the other will not be similar until they are blurred using a large Gaussian kernel or shrunk to small dimensions. This observation enables us to roughly tell which objects in the scene are closer to us and which ones are farther away. At the end of this paper, we will discuss the limitations of our proposed approaches and some of the directions for future work.

Recent research on magnitude Magnetic Resonance Images (MRI) reconstruction from the Fourier inverse transform of complex (gaussian contaminated) data sets focuses on the proper modeling of the resulting Rician noise contaminated data. In this paper we consider a variational Rician denoising model for MRI data sets that we solve by a semi-implicit numerical scheme, which leads to the resolution of a sequence of Rudin, Osher and temi (ROF) models. The (iterated) resolution of these well posed numerical problems is then proposed for Total Variation (TV) Rician denoising. For numerical comparison we also consider a direct semi-implicit approach for the primal problem which amounts to consider some (regularizing) approximating problems. Synthetic and real MR brain images are then denoised and the results show the effectiveness of the new method in both, the accuracy and the speeding up of the algorithm.

Recently, the digitization of paper-based documents is rapidly advanced through the spread of scanners. However, tagging or sorting a huge amount of scanned documents one by one is difficult in terms of time and effort. Therefore, the system which extracts features from texts in the documents automatically, which is available by OCR, and classifies/retrieves documents will be useful. LDA, one of the most popular Topic Models, is known as a method to extract the features of each document and the relationships between documents. However, it is reported that the performance of LDA declines along with poor OCR recognition. This paper assumes the case of applying LDA to Japanese OCR documents and studies the method to improve the performance of topic inference. This paper defines the reliability of the recognized words using N-gram and proposes the weighting LDA method based on the reliability. Adequacy of the reliability of the recognized words is confirmed through the preliminary experiment detecting false recognized words, and then the experiment to classify practical OCR documents are carried out. The experimental results show the improvement of the classification performance by the proposed method comparing with the conventional methods.

The positioning accuracy with Global Navigation Satellite System (GNSS) depends on both the pseudorange error and the geometric dilution of precision (GDOP), and GDOP is depended on the geometry of user/satellites which are selected. Positioning with a GNSS needs at least four satellites which are visible for the user so that at least four pseudorange measurements can be provided to determine the user position in three dimensions and the receiver clock offset. In order to select as few as visible satellites whose geometry is better, a new fast satellite selection algorithm based on geometry is come up with by this paper through analyzing geometry characteristics of the optimal visible satellites combination. The expected visible satellites are selected based on the distribution characteristics of elevation angles and azimuth angles combined with tetrahedron volume. Taking GPS, GLONASS and GALILEO system for examples, this paper compares the algorithm proposed by this paper with the best geometry satellite selection algorithm through simulation. The simulation results show that the algorithm proposed by this paper solves the problem that there are a lot of matrix multiplications and matrix inversions in the best geometry satellite selection algorithm. So the new algorithm can reduce computational complexity and increase receiver processing speed. The theory of the algorithm is simple and easy to use. Meanwhile, the algorithm can satisfy the real time requirements for users.

In the decades since Markov chain Monte Carlo methods were first introduced, they have revolutionised Bayesian approaches to statistical inference. Each new advance in MCMC methodology produces near immediate benefits for Bayesian practitioners, expanding the range of problems they can feasibly solve. In this paper, we explore ways in which Bayesian approaches can return something of the debt owed to MCMC, by using explicitly Bayesian concepts to aid in the design of MCMC samplers. The art of efficient MCMC sampling lies in designing a Markov process that (a) has the required limiting distribution, (b) has good convergence and mixing properties and (c) can be implemented in a computationally efficient manner. In this paper, we explore the idea that the selection of an appropriate process, and in particular the tuning of the parameters of the process to achieve the above goals, can be regarded as a problem of estimation. As such, it is amenable to a conventional Bayesian approach, in which a prior distribution for optimal parameters of the sampler is specified, data relevant to sampler performance is obtained and a posterior distribution for optimal parameters is formed. Sampling from this posterior distribution can then be incorporated into the MCMC sampler to produce an adaptive method. We present a new MCMC algorithm for Bayesian adaptive Metropolis-Hasting sampling (BAMS), using an explicitly Bayesian inference to update the proposal distribution. We show that author Keith’s earlier Bayesian adaptive independence sampler (BAIS) and a new Bayesian adaptive random walk sampler (BARS) emerge as instances. More important than either of these instances, BAMS provides a general framework within which to explore adaptive schemes that are guaranteed to converge to the required limiting distribution.

To digest tremendous documents efficiently, people often resort to their titles, which normally provide a concise and semantic representation of main text. Some titles however are misleading due to lexical ambiguity or eye-catching intention. The requirement of reference summaries hampers using traditional lexical summarisation evaluation techniques for title evaluation. In this paper we develop semantic title evaluation techniques by comparing a title with other sentences in terms of topic-based similarity with regard to the whole document. We further give a statistical hypothesis test to check whether a title is favourable without any reference summary. As a byproduct, the top similar sentence can be recommended as a candidate for title. Experiments on patents, scientific papers and DUC’04 benchmarks show our Semantic Title Evaluation and Recommendation technique based on a recent Segmented Topic Model (STERSTM), performs substantially better than that based on the canonical model Latent Dirichlet Allocation (STERLDA). It can also recommend titles with quality comparable with the winners of DUC’04 in terms of summarising documents into very short summaries.

This year marks the 40th anniversary of Charles Bennett's seminal paper on reversible computing. Bennett's contribution is remembered as one of the first to demonstrate how any deterministic computation can be simulated by a logically reversible Turing machine. Perhaps less remembered is that the same paper suggests the use of nucleic acids to realise physical reversibility. In context, Bennett's foresight predates Leonard Adleman's famous experiments to solve instances of the Hamiltonian path problem using strands of DNA -- a landmark date for the field of natural computing -- by more than twenty years. The ensuing time has seen active research in both reversible computing and natural computing that has been, for the most part, unrelated. Encouraged by new, experimentally viable DNA computing models, there is a resurgent interest in logically reversible computing by the natural computing community. We survey these recent results, and their underlying ideas, which demonstrate the potential for logically and physically reversible computation using nucleic acids.

This paper adapts a novel model, deep belief nets (DBN), to extract entity mentions in Chinese documents. Our experiments were designed to develop entity detection system and entity categorization system applying DBN, to complete entity extraction. Our results exhibit how the depth of architecture and quantity of unit in hidden layer of DBN influence the performance. In DBN Systems, token labels are produced independently and DBN does not concerned what the labels of surrounding token are. Viterbi algorithm is a good solution to overcome this issue. It can find the most likely probability label path to make DBN to be more suitable for entity detection. Furthermore, this paper demonstrates DBN is proper model for our tasks and its results are better than Support Vector Machine (SVM), Artificial Neural Network (ANN) and Conditional Random Field (CRF).

In this paper, we present a scheme for the stability analysis of autonomously controlled production networks with transportations. We model production networks by differential equations and discrete event simulation models (DES) from a mathematical and engineering point of view, where transportation times are considered in the models as time delays. Lyapunov functions as a tool to check the stability of networks are used to calculate stability regions. Then, this region is refined using the detailed DES. This approach provides a scheme to determine stability regions of networks with less time consumption in contrast to a pure simulation approach. In presence of time delays, new challenges in the analysis occur, which is pointed out in this paper.

The paper introduces circular rotation testing system by which GPS receiver’s dynamic performance can be examined, analyzes the change rule of Doppler frequency shift caused by the circular rotation of antenna, compares the states of antenna circular rotation and the linear moving of airborne by investigating the Doppler shift and Doppler shift changing rate, then puts forward the basic method for detecting receiver dynamic positioning accuracy via antenna circular rotation to simulate the state of the airborne linear motion by investigating Doppler shift changing rate. As an example, by using a NovAtel RT2 GPS receiver, the paper gives accuracy analysis of the GPS receiver point positioning and carrier phase differential positioning under the condition of antenna circular rotation with the proposed method, and gives the Receiver’s accuracy of point positioning and carrier phase differential positioning under different moving states as well. Finally, points out the application and popularization of the proposed method in detecting dynamic accuracy and tracking performance of Compass and GPS of other types.

Recently, image classification has been an active research topic due to the urgent need to retrieve and browse digital images via semantic keywords. Based on the success of low-rank matrix recovery which has been applied to statistical learning, computer vision and signal processing, this paper presents a novel low-rank matrix recovery algorithm with discriminant regularization. Standard low-rank matrix recovery algorithm decomposes the original dataset into a set of representative basis with a corresponding sparse error for modeling the raw data. Motivated by the Fisher criterion, the proposed method executes low-rank matrix recovery in a supervised manner, i.e., taking the with-class scatter and between-class scatter into account when the whole label information is available. The paper shows that the formulated model can be solved by the augmented Lagrange multipliers, and provide additional discriminating ability to the standard low-rank models for improved performance. The representative bases learned by the proposed method are encouraged to be structural coherence within the same class, and as independent as possible between classes. Numerical simulations on face recognition tasks demonstrate that the proposed algorithm is competitive with the state-of-the-art alternatives.

In this paper, we compare and analyze the novelty of a scientific paper (text document) of a specific domain. Our experiments utilize the standard Latent Dirichlet Allocation (LDA) topic modeling algorithm to filter the redundant documents and the Ontology of a specific domain which serves as the knowledge base for that domain, to generate cognitive maps for the documents. We report results based on the distance measure such as the Euclidean distance measure that analyses the divergence of the concepts between the documents.

It is a challenging topic how to achieve the real-time tracking of fast video object under complex environment. In this paper, a scheme and its corresponding implementing algorithms of real-time tracking of fast video object are designed and perfected, which are characterized with high performances of real-time tracking and robustness. At first, a kind of scheme is designed for the real-time tracking of fast video object and corresponding implementing strategies for some key modules are proposed. Then the particle filter is employed to predict the pose state of fast video object and the motion object is discriminated from its background by Gaussian kernel and random projection. Moreover, an adaptive feature selection method is used to enhance the robustness and tracking efficiency. A series of experiment results demonstrate that the scheme and algorithms proposed in this paper outperform the current existing algorithms in the tracking efficiency, accuracy and robustness.

It is well-known that the Dolev-Yao adversary is a powerful adversary. Besides acting as the network, intercepting, sending, and composing messages, he can remember as much information as he needs. That is, his memory is unbounded. We recently proposed a weaker Dolev-Yao like adversary, which also acts as the network, but whose memory is bounded. We showed that this Bounded Memory Dolev-Yao adversary, when given enough memory, can carry out many existing protocol anomalies. In particular, the known anomalies arise for bounded memory protocols, where there is only a bounded number of concurrent sessions and the honest participants of the protocol cannot remember an unbounded number of facts nor an unbounded number of nonces at a time. This led us to the question of whether it is possible to infer an upper-bound on the memory required by the Dolev-Yao adversary to carry out an anomaly from the memory restrictions of the bounded protocol. This paper answers this question negatively (Theorem 2). The second contribution of this paper is the formalization of Progressing Collaborative Systems that may create fresh values, such as nonces. In this setting there is no unbounded adversary, although bounded memory adversaries may be present. We prove the NP-completeness of the reachability problem for Progressing Collaborative Systems that may create fresh values.

This paper proposes to add the multithreaded Graphic Processing Units (GPUs) to some virtual machines (VMs) in the existing cloud-based VM groups. To handle the multidimensional or multithreaded computing that a CPU cannot process quickly by a GPU that has hundreds of Arithmetic Logic Units (ALUs), and to regulate the time for initiating physical servers by real-time thermal migration, our proposed scheme can enhance the system performance and reduce the energy consumption of long-term computing. Four major techniques in this paper include: (1) GPU virtualization, (2) Hypervisor for GPU, (3) Thermal migration implementation, and (4) Estimation of multithreaded tasks. In no matter quantum mechanics, astronomy, fluid mechanics, or atmospheric simulation and prediction, a GPU suits not only parallel multithreaded computing for its tens of times performance than a CPU, but also multidimensional array operations for its excellent efficiency. Therefore, how to distribute the computing performance of CPUs and GPUs appropriately becomes a significant issue. In general cloud computing applications, it is rarely seen that GPUs can outperform CPUs. Furthermore, for groups of virtual servers, many tasks actually can be completed by CPUs without the support of GPUs. Thus, it is a waste of resources to implement GPUs to all physical servers. For this reason, by integrating with the migration characteristic of VMs, our proposed scheme can estimate whether to compute tasks by physical machines with GPUs or not. In estimating tasks, we use Amdahl’s law to estimate the overall performance include communication delays, Synchronization overhead and me possible additional burden.

The issue of quantum states’ transfer – in particular, for so-called Perfect State Transfer (PST) – in the networks represented by the spin chains seems to be one of the major concerns in quantum computing. Especially, in the context of future communication methods that can be used in broadly defined computer science. The paper presents a definition of Hamiltonian describing the dynamics of quantum data transfer in one-dimensional spin chain, which is able to transfer the state of unknown qudits. The main part of the paper is the discussion about possibility of entangled states’ perfect transfer, in particular, for the generalized Bell states for qudits. One of the sections also contains the results of numerical experiments for the transmission of quantum entangled state in a noisy quantum channel.

Many countries attach importance to maritime surveillance for its wide applications in exclusive economic zone surveillance, environmental monitoring and anti-piracy operations, etc. Space-borne Synthetic Aperture Radar (SAR) is optimal for its high resolution over wide swaths and all weather working capabilities for ship surveillance, but limited to the level of SAR imaging and image interpretation. Meanwhile, with rapid development of space-borne Automatic Identification System (AIS), almost real time and global coverage for ship surveillance has become possible, but not all ships operate or equip with AIS terminal transmitter. Space-borne SAR and AIS has cooperative nature, so ship surveillance by integration of them has attracted more attention. This paper focuses on the data association of space-borne SAR and AIS. State-of-the-art association methods have good performance except in the high-sea-state and high-density-shipping situations. To improve the flexibility and validity, we emphasis on analyzing two factors: AIS-projected position and Doppler displacement, which mainly affect the accuracy. Firstly, this paper presents the theory of data association in detail. Then state-of-the-art method for projecting AIS position based on Dead Reckoning is researched on further, and improved method based on Gray Prediction Model MGRM (1, N) is introduced. High-accuracy estimation of Doppler displacement is discussed later profoundly. The application of Point Pattern Matching using Shape Context in the association is also investigated, and the simulation results illustrate that it outperforms current state-of-the-art method both in the precision as well as robustness.

Lyapunov exponents of a dynamical system give information about its long-term evolution. Exponents estimation is not an easy task; it is computationally costly and, in presence of chaotic dynamics, it exhibits numerical difficulties. In a previous paper, the authors proposed an algorithm for Lyapunov exponents estimation in piecewise linear systems that strongly reduces the computational time. In this paper, the algorithm is applied also to chaotic systems with polynomial nonlinearity. Firstly, a suitable piecewise linear approximation for the polynomial nonlinear function is evaluated by means of a Multi-Layer Perceptron neural network with linear and saturating linear transfer functions. Then, the linearity of the new state equation and of the variational equation, obtained resorting to the piecewise linear approximation of the nonlinear function, are exploited to accurately evaluate Lyapunov exponents of the approximated system with a reduced execution time.

In this paper the differential kinematics for cable-driven robots is derived and the use for calibration, system investigation and a force based forward kinematics is shown. The Jacobians for each part of the kinematic chain are derived with respect to the platform pose and the most important system parameters. Beside the consideration of geometrical quantities, the differential relations between non-geometrical quantities such as cable stiffness and cable forces are determined. The decomposition in the most fundamental Jacobians allows to analyse and compute more complex relations by reassembling the Jacobians as needed. This approach allows more insight in the system behavior and enables the reuse of the individual modules. The purpose of this paper is to provide the framework and the key equations and to show the use for calibration, force based forward kinematics and system analysis as well as for control purposes.