Norm-based Model Research Articles

Generalized temporal similarity-based nonnegative tensor decomposition for modeling transition matrix of dynamic collaborative filtering

In the real world, user preferences change dynamically. Therefore, time-aware recommendation systems have attracted more attention in both academia and industry. In the literature, tensor decomposition-based models and matrix factorization-based models can handle large-scale sparse data well. However, to the best of our knowledge, there is no work that provides an explanation of the latent time factor embedded in the models. Moreover, conventional Frobenius norm-based models cannot well describe the dynamic changes in user preferences over time. To capture the dynamic changes in user preferences, we interpret the time latent factor vector as a transition matrix of user preferences. In addition, a novel temporal similarity measure is proposed accordingly, which considers dynamic user and item changes between two adjacent time slices. Moreover, we propose a generalized temporal similarity-based nonnegative tensor decomposition (GTS-NTD) model and provide the corresponding solution method. Experiments on three datasets suggest that our proposed method can improve recommendation performance under dynamic changes in user preferences.

The cordon sanitaire: a social norm-based model

ABSTRACT This paper is on cordons sanitaires – coordinated refusals to collaborate with a given political party on principle – and the behaviours associated with it. I argue that, once established, the cordon is properly classified and analysed as a social meta-norm, rather than a mere strategy (the common classification in the literature). First, I examine the merits of a mere strategy model, and conclude that a cordon may well be instituted for strategic reasons, but that aspects of party behaviour are not well accounted for within this model. Second, I show that once established, the cordon is best classified as a social norm, and I discuss the empirical and theoretical implications of this. Finally, I illustrate these theoretical points, through the example of the isolated Sweden Democrats (SD) at the local level. Conceptualizing the cordon sanitaire as a social norm, rather than as a strategy gives new insights into the conditions under which it is complied with, violated, and enforced through sanctions. It also provides an explanation for previously puzzling outcomes, such as the failure of mainstream parties to abandon the cordon once it proves ineffective in limiting the challenger’s growth and influence.

A Robust Multiframe Image Super-Resolution Method in Variational Bayesian Framework

Multiframe image super-resolution (MISR) combines complementary information of a set of low-resolution (LR) images to reconstruct a high-resolution (HR) one. In this study, we propose a robust and fully data-driven MISR method in the variational Bayesian framework. Different from the existing variational super-resolution (SR) methods, we use the l1 norm-based observation model, which takes the acquisition noise, outliers, and impulse noise into account. Furthermore, we have evaluated three typical image prior models, and the most appropriate one is chosen for our proposed method. The proposed method has the following advantages: (1) the HR image and all parameters are automatically estimated in an optimal stochastic sense; (2) the algorithm is robust to impulse noise and outliers. Extensive experiments with synthetic and real images demonstrate the advantages of the proposed method.

Adaptive blind deconvolution using generalized cross-validation with generalized lp/lq norm regularization

Blind image deconvolution is a typically ill-conditioned inverse problem that requires additional information to constrain the solution space. The purpose of this paper is to investigate the characteristics of generalized lp/lq norm on the derivatives of nature images and present a novel efficient method for blind image deconvolution with generalized lp/lq norm (0 < p ≤ 1, p < q). Firstly we analyze the mathematical characteristics of generalized lp/lq norm base on the BSDS dataset, then apply generalized lp/lq norm-based prior model in gradient space to estimate the blur kernel. Due to the complexity of optimization model, we develop an alternating gradient descent method to solve the generalized lp/lq norm-based model which can achieve high recovery quality. Specifically, the selection strategy of regularization parameters is given by using generalized cross-validation method, and these parameters can be updated in alternating minimization steps. Our preliminary experiments show that the proposed method can achieve state-of-the-art results.

Brain Medical Image Fusion Using L2-Norm-Based Features and Fuzzy-Weighted Measurements in 2-D Littlewood–Paley EWT Domain

Computational imaging provides comprehensive and reliable information about human tissue for medical diagnosis and treatment, with medical image fusion as one of the most important technologies in the field. Empirical mode decomposition (EMD), a promising model for image processing, has been used for image fusion in some methods. However, the varying number of decomposed layers leads to problems using EMD for image fusion. In this article, we propose a fusion method for medical images incorporating $L2$ -norm-based features, a match/salience/fuzzy-weighted measure, and the 2-D Littlewood–Paley empirical wavelet transform (2-D LPEWT) as new version of EMD. We first decompose medical images with LPEWT to obtain the residual component (residue) and detailed sub-images that are named as intrinsic mode functions (IMFs). Then we extract the regional features of residue with an $L2$ -norm-based model to fuse the residue while simultaneously fusing IMFs using a method combining a fuzzy membership function with a match/salience measurement. Finally, we reconstruct the comprehensive image by applying inverse LPEWT to the fused residue and IMFs. We evaluated our method using a frequently-used data set of brain images. The results show that our proposed method is more effective than conventional methods by fusing more information into the final images. We also show a feasible scheme for applying EMD to image fusion.

Text Image Deblurring Using Kernel Sparsity Prior.

Previous methods on text image motion deblurring seldom consider the sparse characteristics of the blur kernel. This paper proposes a new text image motion deblurring method by exploiting the sparse properties of both text image itself and kernel. It incorporates the L 0 -norm for regularizing the blur kernel in the deblurring model, besides the L 0 sparse priors for the text image and its gradient. Such a L 0 -norm-based model is efficiently optimized by half-quadratic splitting coupled with the fast conjugate descent method. To further improve the quality of the recovered kernel, a structure-preserving kernel denoising method is also developed to filter out the noisy pixels, yielding a clean kernel curve. Experimental results show the superiority of the proposed method. The source code and results are available at: https://github.com/shenjianbing/text-image-deblur.

Parallel three-dimensional electrical capacitance data imaging using a nonlinear inversion algorithm and [formula omitted] norm-based model regularization

In order to improve image reconstructions, different classes of nonlinear inversion algorithms are developed and used in different research topics like imaging processes in oil industry or the characterization of complex porous media or multiphase flows. These algorithms are able to avoid local minima and to reach more adapted minima of a given misfit function between observed/measured and computed data. Techniques as different as electrical, ultrasound or potential methods, are used. We present here a nonlinear algorithm that allows us to produce permittivity images by using electrical capacitance tomography (ECT). ECT is a non-invasive technique to image non-conductive permittivity distributions and is used in many oil industry imaging applications such as multiphase flows in pipelines, fluidized bed reactors, mixing vessels, and tanks of phase separation. Even if the ECT technique provides low resolution reconstructions, it is cheap, robust and very fast when compared to other imaging tools. In this method one or more rings of electrodes excite a medium to be imaged at high frequencies, and more particularly at frequencies for which a static electrical potential field has fully developed. In many studies of other research groups only one ring of sources is introduced but the reconstruction accuracy was not totally satisfactory due to the 3D nature of the problem to be solved. Instead of using nonlinear stochastic algorithms like the simulated annealing (SA) technique that we optimized in previous studies to image permittivity distributions of granular or solid materials as well as real oil–gas or two-phase flows in 2D cylindrical vessel configurations, we propose here a new ECT inversion tool to image permittivities in a 3D cylindrical configuration. 3D stochastic optimization methods such as SA, neural networks, genetic algorithms can become computationally too prohibitive, and classical local or linear inversion methods excessively smooth images in many cases. Therefore, we propose here a 3D parallel inversion procedure with different numbers of rings and different Lp norms, with1<p⩽2, applied to the model regularization of the misfit function to increase the resolution of the models after inversion. We are able to better reconstruct two-phase and three-phase (oil, gas and solids) mixtures by combining Lp-norm regularizations of the misfit function to minimize and several rings of electrodes. All these algorithms have been implemented in a more general parallel framework TOMOFAST-X designed for multi-physics joint inversion purposes, and could also be used in other fields of research such as larger-scale geophysical exploration for instance.

Beyond opponent coding of facial identity: Evidence for an additional channel tuned to the average face.

Face identity can be represented in a multidimensional space centered on the average. It has been argued that the average acts as a perceptual norm, with the norm coded implicitly by balanced activation in pairs of channels that respond to opposite extremes of face dimensions (two-channel model). In Experiment 1 we used face identity aftereffects to distinguish this model from a narrow-band multichannel model with no norm. We show that as adaptors become more extreme, aftereffects initially increase sharply and then plateau. Crucially there is no decrease, ruling out narrow-band multichannel coding, but consistent with a two-channel norm-based model. However, these results leave open the possibility that there may be a third channel, tuned explicitly to the norm (three-channel model). In Experiment 2 we show that alternating adaptation widens the range identified as the average whereas adaptation to the average narrows the range, consistent with the three-channel model. Explicit modeling confirmed the three-channel model as the best fit for the combined data from both experiments. However, a two-channel model with decision criteria allowed to vary between adapting conditions, also provided a very good fit. These results support opponent, norm-based coding of face identity with additional explicit coding of the norm. (PsycINFO Database Record

Not just the norm: exemplar-based models also predict face aftereffects.

The face recognition literature has considered two competing accounts of how faces are represented within the visual system: Exemplar-based models assume that faces are represented via their similarity to exemplars of previously experienced faces, while norm-based models assume that faces are represented with respect to their deviation from an average face, or norm. Face identity aftereffects have been taken as compelling evidence in favor of a norm-based account over an exemplar-based account. After a relatively brief period of adaptation to an adaptor face, the perceived identity of a test face is shifted toward a face with attributes opposite to those of the adaptor, suggesting an explicit psychological representation of the norm. Surprisingly, despite near universal recognition that face identity aftereffects imply norm-based coding, there have been no published attempts to simulate the predictions of norm- and exemplar-based models in face adaptation paradigms. Here, we implemented and tested variations of norm and exemplar models. Contrary to common claims, our simulations revealed that both an exemplar-based model and a version of a two-pool norm-based model, but not a traditional norm-based model, predict face identity aftereffects following face adaptation.

Robust compressive multi‐input–multi‐output imaging

Canonical multi-input–multi-output (MIMO) imaging methods suffer from limited resolution, poor robustness against noise and high computational complexity, especially when the array aperture is limited (affecting angular resolution) and the number of snapshots is limited (affecting Doppler resolution). In this study, the authors discuss a new range-angle-Doppler monostatic MIMO imaging method through adaptive estimation of the generalised Cauchy prior distribution (GCD). The superiority of GCD-based model over existing l p -norm-based model (which actually assumes the prior as general Gaussian distribution) is theoretically verified through the issue of signal compressibility. The authors adapt a reweighted l2-norm iterative algorithm to solve the model. In our model the authors do not pre-define the scaling factor of the prior distribution and, during the iteration, the authors use a novel ‘quantile-of-OS’ method to adaptively estimate the scaling parameter of the prior distribution, enhancing the robustness of the method. Simulation results verify the image quality and speed advantages of the proposed method.

Students’ Expectations About Their Grades Versus Course Expectations From Them

The paper presents a research study on the reasons for mismatch between students’ grade expectations and the actual grades they were awarded in the “Ethics” course at an engineering education institution in India. The paper also outlines how this event created the social and political stigmas among the authorities to opt for the re-evaluation of students’ answer sheets. After receiving the re-evaluated grade sheet from the re-evaluator, some meager and minor discrepancies were noticed which were liable to question, suspicion, and hence not impeccable. But by overlooking the reasons for trifling discrepancies authorities considered the re-evaluated grade sheet as the final one. In this context, the study examined the reliability and viability of criteria-based grading against norm-based grading model with the plausible impacts of authoritarian intervention in distorting the course instructor’s grade sheet, and its implications on quality retention in higher education.

The Road Not Taken

To investigate individual differences in creativity as measured with a complex problem-solving task, we developed a computational model of the remote associates test (RAT). For 50 years, the RAT has been used to measure creativity. Each RAT question presents three cue words that are linked by a fourth word, which is the correct answer. We hypothesized that individuals perform poorly on the RAT when they are biased to consider high-frequency candidate answers. To assess this hypothesis, we tested individuals with 48 RAT questions and required speeded responding to encourage guessing. Results supported our hypothesis. We generated a norm-based model of the RAT using a high-dimensional semantic space, and this model accurately identified correct answers. A frequency-biased model that included different levels of bias for high-frequency candidate answers explained variance for both correct and incorrect responses. Providing new insight into the nature of creativity, the model explains why some RAT questions are more difficult than others, and why some people perform better than others on the RAT.

Face adaptation: Comparing norm-based and exemplar-based models

Face adaptation: Comparing norm-based and exemplar-based models

A model of face space

Faces vary in their distinctiveness, and it has been suggested that—more distinctive faces are located further from the mean face in internal or perceptual face space (Valentine, 1991), than typical faces. This paper proposes a pixel-based computational model of face distinctiveness. I show that the notion of psychological distinctiveness is correlated with the metrical Euclidean distance, from the mean face. Distances are measured in pixel-based face space, which serves as the basis of the computational model. A prediction of the model is that pairwise perceptual distances between distinctive faces are, on average, greater than between typical faces. This is affirmed by psychophysical experiment. This finding confirms a key axiom of the norm-based coding model, which states that face space is centrally distributed with typical faces packed densely near the centre of the space and distinctive faces spread sparsely in the periphery. This assumption has previously been confirmed by Davidenko (2007) using face silhouettes, and here I show that this axiom also generalizes to full frontal face images.

LIVING LIKE THE LAITY? THE NEGOTIATION OF RELIGIOUS STATUS IN THE CITIES OF LATE MEDIEVAL ITALY

ABSTRACTFramed by consideration of images of treasurers on the books of the treasury in thirteenth-century Siena, this article uses evidence for the employment of men of religion in city offices in central and northern Italy to show how religious status (treated as a subset of ‘clerical culture’) could become an important object of negotiation between city and churchmen, a tool in the repertoire of power relations. It focuses on the employment of men of religion as urban treasurers and takes Florence in the late thirteenth and early fourteenth centuries as a principal case study, but also touches on the other tasks assigned to men of religion and, very briefly, on evidence from other cities (Bologna, Brescia, Como, Milan, Padua, Perugia and Siena). It outlines some of the possible arguments deployed for this use of men of religion in order to demonstrate that religious status was, like gender, more contingent and fluid than the norm-based models often relied on as a shorthand by historians. Despite the powerful rhetoric of lay–clerical separation in this period, the engagement of men of religion in paid, term-bound urban offices inevitably brought them closer to living like the laity.

Changing faces: Direction is important

Recent research has provided evidence for the role of norm-based coding in face recognition (e.g., Leopold et al., 2001). In such a model, any given face can be represented by a vector from the norm, with the difference between two faces being the difference between the two vectors. However, as has previously been suggested it is also possible to conceive of a model in which the angle between two vectors provides relevant information for differentiating faces. Two experiments investigated this possibility using a face matching paradigm. Discrimination was found to be better when two faces lay on different vectors than when they were the same physical distance apart but lying on the same vector. The results favour a specific type of norm-based model suggesting that vector angle is important in face perception.

How is the individuality of a face recognized?

A familiar face is instantly recognized in a crowd. This cannot be achieved through a feature by feature comparison of the observed face with either an average face (norm-based model of face recognition) or with a set of similarly constructed faces stored in memory (exemplar-based model of face recognition). A modified norm-based model is thus proposed. Instead of memorizing an average face, the normal variations for each facial feature are used to construct a multidimensional volume of face-space devoid of unusual features, here defined as features whose metrics lie below the 5th or above the 95th percentiles for that feature. A face consisting of 100 independently variable features will thus have, on average, 10 unusual features. Face identification then becomes exception-reporting. It requires only 10 such rare features to render a given face a one in 10 13 faces ( P=0.05 10=9.8×10 −14). In a world containing 6.7×10 9 people, such a face would be unique. Faces remembered in this way can have their unusual features exaggerated or attenuated without loss of identity. This is the basis of caricatures and anti-caricatures. It also means that individuals belonging to a foreign race, possessing several features with modes beyond the “usual range” of the own-race population, will all look alike. Features that render a face unique in the own-race population are now shared by everyone in the foreign race. Average faces are more beautiful than the faces used in the averaging process. This makes evolutionary sense. Natural selection increases the frequency of fit features at the expense of maladaptive features. “Usual features” are therefore fitter than “unusual features”, and play an important role in mate selection. Such an existing fundamental sexual attribute could easily have been harnessed for the fast and efficient recognition of individuals in the community.

Aftereffects for face attributes with different natural variability: Adapter position effects and neural models.

Adaptation to distorted faces is commonly interpreted as a shift in the face-space norm for the adapted attribute. This article shows that the size of the aftereffect varies as a function of the distortion level of the adapter. The pattern differed for different facial attributes, increasing with distortion level for symmetric deviations of eye height and decreasing for asymmetric deviations. These results are interpreted in terms of different coding ranges for the 2 facial attributes, arising from differences in eye-height variability in natural face images (large for symmetric, small for asymmetric). Neural models developed in low-level vision also are applied to facial attributes, contrasting a 2-pool (norm-based) and a multichannel (exemplar-based) model. The adapter position effects generally support a norm-based model, as did a finding that perception of stimuli further from the norm than the adapter was shifted in the direction of the norm, rather than repulsed away from the adapter.

Models of faces representation

Considering that information alteration of faces and other visual stimulus differ qualitative, it is important to establish models, which would enable to explain peculiar consistent patterns of information processing of faces.One of the most important stages of informatikon processing – representation in memory. Recently, seeking to discover the peculiarities of faces representation in memory, the biggest attention is given to face space of T. Valentine and patterns of M. B. Lewis and R. A. Johnston Voronoi.In T. Valentine’s multidimensional face-space framework, faces are encoded according to a series of visual or physical parameter particular to each face. Each of these parameters is represented along a space dimension. The dimensions of the space will be those that serve to discriminate between faces and they represent the perceptual dimensions or features of faces that distinguish the faces. A face’s representation will be made up of many values taken along many different dimensions. This set of values will be unique for every individual face and will be represented as a single point (or vector) in an N-dimensional face-space. The main problem – because of non-comprehensive description of face’s dimension and encoded information of faces it is not clear the way of distinguishing physical and psychological faces’ dimensions, the possibilities to range the encoded face dimensions according to their importance and significance.Two different models based on this framework, norm-based and exemplar-based models are described and contrasted. Models that incorporate the facial prototype are referred to as norm-based models. On such models, faces are represented by face vectors taken relative to the facial prototype at the center of the face-space, which represent the average value of the population on each dimension. Although the existence of face-prototype is certified by the research of faces’ caricature, it is unclear why the faces of representatives of other races are reproduced inaccurate, how are the valuation mediums of encoded faces’ dimension calculated, does the face-prototype is kept in memory as the real face?As an alternative to these – exemplar-based models have face representations that are stored as absolutes, relative only to other exemplars. The present study describes Voronoi diagram, developed from the exemplar-based model, which accounts for the caricature advantage without reference to an explicitly encoded facial prototype. Voronoi diagram involves interpreting the face-space as a space of regions of identity (also called „Voronoi cells“) rather than a set of points or vectors. The advantage of this pattern – the identity of observed face is connected with appropriate „cell“. It is important, because such structure, differently to the pattern oriented to standard, enables to explain the fact, that faces are recognized despite their dimensions or change of circumstances of observation.Distinguishing between these two models remains difficult because they make similar predictions. For example, both models predict that faces that are judged to be distinctive will be recognized faster and more accurately than those judged to be typical. Also indicate that typical faces are generally located near the center of the space in more densely populated facespace while distinctive faces will tend to be present in the outer regions. However, with help of patterns oriented to standards, as well as to examples, it is possible to explain only some of peculiarities of informatikon alteration of faces and the results of empirical researches are quite contradictory. Further analysis of pattern of faces’ expansion is necessary – it would not only specify the hypothetical structure of faces’ expansion, but also elaborate the proceeding processes.

A Norm-Based Remedial Model for Underinclusive Statutes

A Norm-Based Remedial Model for Underinclusive Statutes