Abstract: In this paper we describe a novel technique for detecting salient regions in an image. The detector is a generalization to affine invariance of the method introduced by Kadir and Brady [10]. The detector deems a region salient if it exhibits unpredictability in both its attributes and its spatial scale.

Abstract: Discriminant analysis has been used for decades to extract features that preserve class separability. It is commonly defined as an optimization problem involving covariance matrices that represent the scatter within and between clusters. The requirement that one of these matrices be nonsingular limits its application to data sets with certain relative dimensions. We examine a number of optimization criteria, and extend their applicability by using the generalized singular value decomposition to circumvent the nonsingularity requirement. The result is a generalization of discriminant analysis that can be applied even when the sample size is smaller than the dimension of the sample data. We use classification results from the reduced representation to compare the effectiveness of this approach with some alternatives, and conclude with a discussion of their relative merits.

Abstract: The well-known privacy-preserved data mining modifies existing data mining techniques to randomized data. In this paper, we investigate data mining as a technique for masking data, therefore, termed data mining based privacy protection. This approach incorporates partially the requirement of a targeted data mining task into the process of masking data so that essential structure is preserved in the masked data. The idea is simple but novel: we explore the data generalization concept from data mining as a way to hide detailed information, rather than discover trends and patterns. Once the data is masked, standard data mining techniques can be applied without modification. Our work demonstrated another positive use of data mining technology: not only can it discover useful patterns, but also mask private information. We consider the following privacy problem: a data holder wants to release a version of data for building classification models, but wants to protect against linking the released data to an external source for inferring sensitive information. We adapt an iterative bottom-up generalization from data mining to generalize the data. The generalized data remains useful to classification but becomes difficult to link to other sources. The generalization space is specified by a hierarchical structure of generalizations. A key is identifying the best generalization to climb up the hierarchy at each iteration. Enumerating all candidate generalizations is impractical. We present a scalable solution that examines at most one generalization in each iteration for each attribute involved in the linking.

Abstract: In this paper we propose a generalization of multiscale finite element methods (Ms-FEM) to nonlinear problems. We study the convergence of the proposed method for nonlinear elliptic equations and propose an oversampling technique. Numerical examples demonstrate that the over-sampling technique greatly reduces the error. The application of MsFEM to porous media flows is considered. Finally, we describe further generalizations of MsFEM to nonlinear time-dependent equations and discuss the convergence of the method for various kinds of heterogeneities.

Abstract: In this paper we generalize the intuitionistic fuzzy set (IFS), paraconsistent set, and intuitionistic set to the neutrosophic set (NS). Several examples are presented. Also, a geometric interpretation of the Neutrosophic Set is given using a Neutrosophic Cube. Many distinctions between NS and IFS are underlined.

Abstract: Biological and technological networks contain patterns, termed network motifs, which occur far more often than in randomized networks. Network motifs were suggested to be elementary building blocks that carry out key functions in the network. It is of interest to understand how network motifs combine to form larger structures. To address this, we present a systematic approach to define "motif generalizations": families of motifs of different sizes that share a common architectural theme. To define motif generalizations, we first define "roles" in a subgraph according to structural equivalence. For example, the feedforward loop triad--a motif in transcription, neuronal, and some electronic networks--has three roles: an input node, an output node, and an internal node. The roles are used to define possible generalizations of the motif. The feedforward loop can have three simple generalizations, based on replicating each of the three roles and their connections. We present algorithms for efficiently detecting motif generalizations. We find that the transcription networks of bacteria and yeast display only one of the three generalizations, the multi-output feedforward generalization. In contrast, the neuronal network of C. elegans mainly displays the multi-input generalization. Forward-logic electronic circuits display a multi-input, multi-output hybrid. Thus, networks which share a common motif can have very different generalizations of that motif. Using mathematical modeling, we describe the information processing functions of the different motif generalizations in transcription, neuronal, and electronic networks.

Abstract: A novel linear matrix inequality (LMI)-based criterion for the global asymptotic stability and uniqueness of the equilibrium point of a class of delayed cellular neural networks (CNNs) is presented. The criterion turns out to be a generalization and improvement over some previous criteria.

Abstract: This paper presents a new approach to estimation and inference in panel data models with unobserved common factors possibly correlated with exogenously given individual-specific regressors and/or the observed common effects. The basic idea behind the proposed estimation procedure is to filter the individual-specific regressors by means of (weighted) cross-section aggregates such that asymptotically as the cross-section dimension (N) tends to infinity the differential effects of unobserved common factors are eliminated. The estimation procedure has the advantage that it can be computed by OLS applied to an auxiliary regression where the observed regressors are augmented by cross sectional averages of the dependent variable and the individual specific regressors. It is shown that the proposed correlated common effects (CCE) estimators for the individual-specific regressors (and its pooled counterpart) are asymptotically unbiased as N approaches infinity, both when T (the time-series dimension) is fixed, and when N and T tend to infinity jointly. A generalization of these results to multi-factor structures is also provided. The estimation and inference in dynamic heterogenous panels with a residual factor structure will be addressed in a companion paper.

Abstract: This letter presents some results of an analysis on the decision boundaries of complex-valued neural networks whose weights, threshold values, input and output signals are all complex numbers. The main results may be summarized as follows. (1) A decision boundary of a single complex-valued neuron consists of two hypersurfaces that intersect orthogonally, and divides a decision region into four equal sections. The XOR problem and the detection of symmetry problem that cannot be solved with two-layered real-valued neural networks, can be solved by two-layered complex-valued neural networks with the orthogonal decision boundaries, which reveals a potent computational power of complex-valued neural nets. Furthermore, the fading equalization problem can be successfully solved by the two-layered complex-valued neural network with the highest generalization ability. (2) A decision boundary of a three-layered complex-valued neural network has the orthogonal property as a basic structure, and its two hypersurfaces approach orthogonality as all the net inputs to each hidden neuron grow. In particular, most of the decision boundaries in the three-layered complex-valued neural network inetersect orthogonally when the network is trained using Complex-BP algorithm. As a result, the orthogonality of the decision boundaries improves its generalization ability. (3) The average of the learning speed of the Complex-BP is several times faster than that of the Real-BP. The standard deviation of the learning speed of the Complex-BP is smaller than that of the Real-BP. It seems that the complex-valued neural network and the related algorithm are natural for learning complex-valued patterns for the above reasons.

Abstract: A fundamental problem in statistical parsing is the choice of criteria and algo-algorithms used to estimate the parameters in a model. The predominant approach in computational linguistics has been to use a parametric model with some variant of maximum-likelihood estimation. The assumptions under which maximum-likelihood estimation is justified are arguably quite strong. This chapter discusses the statistical theory underlying various parameter-estimation methods, and gives algorithms which depend on alternatives to (smoothed) maximum-likelihood estimation. We first give an overview of results from statistical learning theory. We then show how important concepts from the classification literature - specifically, generalization results based on margins on training data - can be derived for parsing models. Finally, we describe parameter estimation algorithms which are motivated by these generalization bounds.

Abstract: The paper describes a probabilistic active learning strategy for support vector machine (SVM) design in large data applications. The learning strategy is motivated by the statistical query model. While most existing methods of active SVM learning query for points based on their proximity to the current separating hyperplane, the proposed method queries for a set of points according to a distribution as determined by the current separating hyperplane and a newly defined concept of an adaptive confidence factor. This enables the algorithm to have more robust and efficient learning capabilities. The confidence factor is estimated from local information using the k nearest neighbor principle. The effectiveness of the method is demonstrated on real-life data sets both in terms of generalization performance, query complexity, and training time.

Abstract: We propose a theory for modeling concepts that uses the state-context-property theory (SCOP), a generalization of the quantum formalism, whose basic notions are states, contexts and properties. This theory enables us to incorporate context into the mathematical structure used to describe a concept, and thereby model how context influences the typicality of a single exemplar and the applicability of a single property of a concept. We introduce the notion `state of a concept' to account for this contextual influence, and show that the structure of the set of contexts and of the set of properties of a concept is a complete orthocomplemented lattice. The structural study in this article is a preparation for a numerical mathematical theory of concepts in the Hilbert space of quantum mechanics that allows the description of the combination of concepts (see quant-ph/0402205)

Abstract: This paper focuses on approaches that address the intractability of knowledge acquisition of conditional probability tables in causal or Bayesian belief networks. We state a rule that we term the "recursive noisy OR" (RNOR) which allows combinations of dependent causes to be entered and later used for estimating the probability of an effect. In the development of this paper, we investigate the axiomatic correctness and semantic meaning of this rule and show that the recursive noisy OR is a generalization of the well-known noisy OR. We introduce the concept of positive causality and demonstrate its utility in axiomatic correctness of the RNOR. We also introduce concepts describing the ways in which dependent causes can work together as being either "synergistic" or "interfering." We provide a formalization to quantify these concepts and show that they are preserved by the RNOR. Finally, we present a method for the determination of Conditional Probability Tables from this causal theory.

Abstract: In the context of solving nonlinear partial differential equations, Shu and Osher introduced representations of explicit Runge-Kutta methods, which lead to stepsize conditions under which the numerical process is total-variation-diminishing (TVD). Much attention has been paid to these representations in the literature. In general, a Shu-Osher representation of a given Runge-Kutta method is not unique. Therefore, of special importance are representations of a given method which are best possible with regard to the stepsize condition that can be derived from them. Several basic questions are still open, notably regarding the following issues: (1) the formulation of a simple and general strategy for finding a best possible Shu-Osher representation for any given Runge-Kutta method; (2) the question of whether the TVD property of a given Runge-Kutta method can still be guaranteed when the stepsize condition, corresponding to a best possible Shu-Osher representation of the method, is violated; (3) the generalization of the Shu-Osher approach to general (possibly implicit) Runge-Kutta methods. In this paper we give an extension and analysis of the original Shu-Osher representation, by means of which the above questions can be settled. Moreover, we clarify analogous questions regarding properties which are referred to, in the literature, by the terms monotonicity and strong-stability-preserving (SSP).

Abstract: The article reviews the book “Validity Generalization: A Critical Review,” edited by Kevin R. Murphy.

Abstract: Support varieties for any finite dimensional algebra over a field were introduced in (20) using graded subalgebras of the Hochschild cohomol- ogy. We mainly study these varieties for selfinjective algebras under appropri- ate finite generation hypotheses. Then many of the standard results from the theory of support varieties for finite groups generalize to this situation. In par- ticular, the complexity of the module equals the dimension of its corresponding variety, all closed homogeneous varieties occur as the variety of some module, the variety of an indecomposable module is connected, periodic modules are lines and for symmetric algebras a generalization of Webb's theorem is true.

Abstract: By generalizing the standard solution for 2-person games into n-person cases, this paper develops a new solution concept for cooperative games: the consensus value.We characterize the consensus value as the unique function that satisfies efficiency, symmetry, the quasi dummy property and additivity.By means of the transfer property, a second characterization is provided.By defining the stand-alone reduced game, a recursive formula for the value is established.We also show that this value is the average of the Shapley value and the equal surplus solution.Furthermore, we discuss a possible generalization.

Abstract: The paper presents recent results concerning the problem of the existence of those selections, which preserve the properties of a given set-valued mapping of one real variable taking on compact values from a metric space. The properties considered are the boundedness of Jordan, essential or generalized variation, Lipschitz or absolute continu- ity. Selection theorems are obtained by virtue of a single compactness argument, which is the exact generalization of the Helly selection prin- ciple. For set-valued mappings with the above properties we obtain a Castaing-type representation and prove the existence of multivalued se- lections and selections which pass through the boundaries of the images of the set-valued mapping and which are nearest in variation to a given mapping. Multivalued Lipschitzian superposition operators acting on mappings of bounded generalized variation are characterized, and so- lutions of bounded generalized variation to functional inclusions and embeddings, including variable set-valued operators in the right hand side, are obtained. Bibliography contains 113 items.

Abstract: We study a simple learning algorithm for binary classification. Instead of predicting with the best hypothesis in the hypothesis class, that is, the hypothesis that minimizes the training error, our algorithm predicts with a weighted average of all hypotheses, weighted exponentially with respect to their training error. We show that the prediction of this algorithm is much more stable than the prediction of an algorithm that predicts with the best hypothesis. By allowing the algorithm to abstain from predicting on some examples, we show that the predictions it makes when it does not abstain are very reliable. Finally, we show that the probability that the algorithm abstains is comparable to the generalization error of the best hypothesis in the class.

Abstract: We present a unified representation theorem for the class of all outer generalized inverses of a bounded linear operator. Using this representation we develop a few specific expressions and computational procedures for the set of outer generalized inverses. The obtained result is a generalization of the well-known representation theorem of the Moore--Penrose inverse as well as a generalization of the well-known results for the Drazin inverse and the generalized inverse AT,S (2). Also, as corollaries we get corresponding results for reflexive generalized inverses.

Abstract: In recent years, methods for the integration of Poisson manifolds and of Lie algebroids have been proposed, the latter being usually presented as a generalization of the former. In this Letter it is shown that the latter method is actually related to (and may be derived from) a particular case of the former if one regards dual of Lie algebroids as special Poisson manifolds. The core of the proof is the fact, discussed in the second part of this Letter, that coisotropic submanifolds of a (twisted) Poisson manifold are in one-to-one correspondence with possibly singular Lagrangian subgroupoids of source-simply-connected (twisted) symplectic groupoids.

Abstract: We shall show that only two components of vorticity play an essential role to determine possibility of extension of the time interval for the local strong solution to the Navier-Stokes equations. Then we shall apply our extension theorem to regularity criterion on weak solutions due to Serrin and Beirao da Veiga. Chae–Choe proved the same criterion as Beirao da Veiga only by means of the two-components of vorticity. We deal with the critical case which they excluded. Our criterion may be regarded as the generalization of the result of Beal-Kato-Majda from L ∞ to BMO.

Abstract: The transmissibility concept may be generalized to multi-degree-of-freedom systems with multiple random excitations. This generalization involves the definition of a transmissibility matrix, relating two sets of responses when the structure is subjected to excitation at a given set of coordinates. Applying such a concept to an experimental example is the easiest way to validate this method.

Abstract: Nonlocality without entanglement is an interesting field. A manifestation of quantum nonlocality without entanglement is the possible local indistinguishability of orthogonal product states. In this paper we analyze the character of operators to distinguish the elements of a full product basis set in a multipartite system, and show that distinguishing perfectly these product bases needs only local projective measurements and classical communication, and these measurements cannot damage each product basis. Employing these conclusions one can discuss local distinguishability of the elements of any full product basis set easily. Finally we discuss the generalization of these results to the locally distinguishability of the elements of incomplete product basis set.