Search

Role negotiation is a pivotal step in the role-based collaboration (RBC) process. There are many factors to be considered when decision makers set up criteria to evaluate group performance and agents’ abilities. In this paper, we investigate the applications of three traditional multicriteria decision making methods into role negotiation of RBC, i.e., simple additive weighting, multiplication exponent weighting, and weighted distance. The goal is to acquire ideal and satisfactory methods to set up the evaluation criteria for the role negotiation task. The main contributions of this paper include: 1) a method of generating the matrix of agent qualifications, with a discrete, usually, limited, and finite set of prespecified attributes and values; 2) the upper and lower thresholds of a role’s ability requirements, which are critical parameters for a workable team and its performance; and 3) the potential influences of specialist agents and generalist agents on the group performance. The proposed methods are verified by simulation experiments. The simulation results show that the proposed methods are reasonable, feasible, and practical. [ABSTRACT FROM AUTHOR]

Reports that Massachusetts Institute of Technology computer science graduate Jeremy Stribling's paper "Rooter: A Methodology for the Typical Unification of Access Points and Redundancy," co-authored with Daniel Aguayo and Maxwell Krohn, was accepted for the 9th World Multi-Conference on Systemics, Cybernetics and Informatics to be held in Florida in July. Development of an automatic computer-science paper generator that cobbles together articles adorned with randomly generated graphs; Other information provided by the paper.

Ensemble clustering has been a popular research topic in data mining and machine learning. Despite its significant progress in recent years, there are still two challenging issues in the current ensemble clustering research. First, most of the existing algorithms tend to investigate the ensemble information at the object-level, yet often lack the ability to explore the rich information at higher levels of granularity. Second, they mostly focus on the direct connections (e.g., direct intersection or pair-wise co-occurrence) in the multiple base clusterings, but generally neglect the multiscale indirect relationship hidden in them. To address these two issues, this paper presents a novel ensemble clustering approach based on fast propagation of cluster-wise similarities via random walks. We first construct a cluster similarity graph with the base clusters treated as graph nodes and the cluster-wise Jaccard coefficient exploited to compute the initial edge weights. Upon the constructed graph, a transition probability matrix is defined, based on which the random walk process is conducted to propagate the graph structural information. Specifically, by investigating the propagating trajectories starting from different nodes, a new cluster-wise similarity matrix can be derived by considering the trajectory relationship. Then, the newly obtained cluster-wise similarity matrix is mapped from the cluster-level to the object-level to achieve an enhanced co-association matrix, which is able to simultaneously capture the object-wise co-occurrence relationship as well as the multiscale cluster-wise relationship in ensembles. Finally, two novel consensus functions are proposed to obtain the consensus clustering result. Extensive experiments on a variety of real-world datasets have demonstrated the effectiveness and efficiency of our approach. [ABSTRACT FROM AUTHOR]

Graphs are commonly used to represent objects, such as images and text, for pattern classification. In a dynamic world, an object may continuously evolve over time, and so does the graph extracted from the underlying object. These changes in graph structure with respect to the temporal order present a new representation of the graph, in which an object corresponds to a set of time-variant graphs. In this paper, we formulate a novel time-variant graph classification task and propose a new graph feature, called a graph-shapelet pattern, for learning and classifying time-variant graphs. Graph-shapelet patterns are compact and discriminative graph transformation subsequences. A graph-shapelet pattern can be regarded as a graphical extension of a shapelet—a class of discriminative features designed for vector-based temporal data classification. To discover graph-shapelet patterns, we propose to convert a time-variant graph sequence into time-series data and use the discovered shapelets to find graph transformation subsequences as graph-shapelet patterns. By converting each graph-shapelet pattern into a unique tokenized graph transformation sequence, we can measure the similarity between two graph-shapelet patterns and therefore classify time-variant graphs. Experiments on both synthetic and real-world data demonstrate the superior performance of the proposed algorithms. [ABSTRACT FROM AUTHOR]

The latent friend recommendation in online social media is interesting, yet challenging, because the user-item ratings and the user–user relationships are both sparse. In this paper, we propose a new dual implicit mining-based latent friend recommendation model that simultaneously considers the implicit interest topics of users and the implicit link relationships between the users in the local topic cliques. Specifically, we first propose an algorithm called all reviews from a user and all tags from their corresponding items to learn the implicit interest topics of the users and their corresponding topic weights, then compute the user interest topic similarity using a symmetric Jensen–Shannon divergence. After that, we adopt the proposed weighted local random walk with restart algorithm to analyze the implicit link relationships between the users in the local topic cliques and calculate the weighted link relationship similarity between the users. Combining the user interest topic similarity with the weighted link relationship similarity in a unified way, we get the final latent friend recommendation list. The experiments on real-world datasets demonstrate that the proposed method outperforms the state-of-the-art latent friend recommendation methods under four different types of evaluation metrics. [ABSTRACT FROM AUTHOR]

Graph matching is a fundamental problem in theoretical computer science and artificial intelligence, and lays the foundation for many computer vision and machine learning tasks. Approximate algorithms are necessary for graph matching due to its NP-complete nature. Inspired by the usage in network-related tasks, random walk is generalized to graph matching as a type of approximate algorithm. However, it may be inappropriate for the previous random walk-based graph matching algorithms to utilize continuous techniques without considering the discrete property. In this paper, we propose a novel random walk-based graph matching algorithm by incorporating both continuous and discrete constraints in the optimization process. Specifically, after interpreting graph matching by random walk, the continuous constraints are directly embedded in the random walk constraint in each iteration. Further, both the assignment matrix (vector) and the pairwise similarity measure between graphs are iteratively updated according the discrete constraints, which automatically leads the continuous solution to the discrete domain. Comparisons on both synthetic and real-world data demonstrate the effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

As the mutation strategy and algorithmic parameters in differential evolution (DE) are sensitive to the problems being solved, a hot research topic is to adaptively control the strategy and parameters according to the requirements of the problem. In the literature, most adaptive DE use either historical experiences of the population or heuristic information of the individuals to promote adaptation. In this paper, we develop a novel variant of adaptive DE, utilizing both the historical experience and heuristic information for the adaptation. In this novel historical and heuristic DE (HHDE), each individual dynamically adjusts its mutation strategy and associated parameters not only by learning from previous successful experience of the whole population, but also according to heuristic information related with its own current state. These help the algorithm select a more suitable mutation strategy and determinate better parameters for each individual in different evolutionary stages. The performance of the proposed HHDE is extensively evaluated on 30 benchmark functions with different dimensions. Experimental results confirm the competitiveness of the proposed algorithm to a number of DE variants. [ABSTRACT FROM AUTHOR]

E-commerce business process nets (EBPNs) are a novel formal model for describing and validating e-commerce systems including interactive parties such as shopper, merchant, and third-party payment platform. Data errors and nondeterminacy of the data states during the trading process can be depicted with the help of EBPNs. However, the problem about how to analyze EBPNs remains largely open. To analyze their data-liveness, data-boundedness, and reachability, this paper presents two analysis methods. For EBPNs, reachability analysis is proposed based on a 3-D incidence matrix method. Additionally, reduction methods are proposed for a special EBPN. Finally, the validity and reliability of the proposed methods are illustrated via the examples of e-commerce systems. [ABSTRACT FROM PUBLISHER]

Linguistic large-scale group decision making (LGDM) problems are more and more common nowadays. In such problems a large group of decision makers are involved in the decision process and elicit linguistic information that are usually assessed in different linguistic scales with diverse granularity because of decision makers’ distinct knowledge and background. To keep maximum information in initial stages of the linguistic LGDM problems, the use of multigranular linguistic distribution assessments seems a suitable choice, however, to manage such multigranular linguistic distribution assessments, it is necessary the development of a new linguistic computational approach. In this paper, it is proposed a novel computational model based on the use of extended linguistic hierarchies, which not only can be used to operate with multigranular linguistic distribution assessments but also can provide interpretable linguistic results to decision makers. Based on this new linguistic computational model, an approach to linguistic large-scale multiattribute group decision making is proposed and applied to a talent selection process in universities. [ABSTRACT FROM AUTHOR]

Scientific practice has been rapidly evolving in the last years under the pressure of developments in computer science and technology. In this paper we present some of the results of our research activity at the boundary between computer science and philosophy of science started in 1997 under Marco Somalvico's impulse and guidance. In particular, we discuss two roles that multiagent systems can play in scientific discovery. From the one hand, they can support scientific practice; from the other hand, they can represent scientific results. The theoretical framework presented in this paper is exemplified in concrete by illustrating specific implemented systems, both taken from the literature and developed by ourselves. [ABSTRACT FROM AUTHOR]

This paper aims to enrich our understanding of the history and substance of cybernetics. It reviews the work of three British cyberneticians — W. Ross Ashby, Stafford Beer and Gordon Pask — paying attention particularly to the materiality of their practice — the strange and fascinating devices and systems that were at the heart of their work — and to the worldly projects they pursued — scientific, technological, artistic, organizational, political and spiritual. Connections are drawn between cybernetics and recent theoretical work in science and technology studies, in the hope of illuminating key features of both. The paper concludes by suggesting that the antidisciplinary impulse of contemporary science studies might find inspiration in the work of cyberneticians — that theory does not have to remain confined to the realm of theory. [ABSTRACT FROM AUTHOR]

This paper focuses on the complexity of the (weak) soundness problem of acyclic workflow (WF) nets, and two main results are established: 1) soundness of 1-bounded acyclic WF nets is co-NP-complete and 2) weak soundness of 3-bounded acyclic asymmetric-choice WF nets is co-NP-complete. [ABSTRACT FROM AUTHOR]

van der Aalst et al. proved that the soundness problem is solvable in polynomial time for free-choice workflow nets (FCWF-nets). However, FCWF-nets cannot model most web services composition and interorganizational business processes because the interaction among processes does not usually satisfy the free-choice requirement. Asymmetric-choice workflow nets (ACWF-nets) as a larger class than FCWF-nets can model lots of such cases. Our previous work showed that the (weak) soundness problem is co-NP-hard for three-bounded ACWF-nets. Later, Tiplea et al. proved that for three-bounded acyclic ACWF-nets, the weak soundness problem is co-NP-complete. We sharp these results in this paper. First, we prove that for ACWF-nets, whether they are one-bounded or k -bounded ( k~\boldsymbol > 1 ), the soundness problem is co-NP-hard. Second, it is proven that the soundness is equivalent to the weak soundness for any acyclic ACWF-nets, i.e., an acyclic ACWF-net is sound if and only if it is weakly sound. [ABSTRACT FROM AUTHOR]

Abstract: The Range Next Value problem (problem RNV) is a recent interesting variant of the range search problems, where the query is for the immediate next (or equal) value of a given number within a given interval of an array. Problem RNV was introduced and studied very recently by Crochemore et al. [Maxime Crochemore, Costas S. Iliopoulos, M. Sohel Rahman, Finding patterns in given intervals, in: Antonin Kucera, Ludek Kucera (Eds.), MFCS, 22 in: Lecture Notes in Computer Science, vol. 4708, Springer, 2007, pp. 645–656]. In this paper, we present improved algorithms for problem RNV and algorithms for extended versions of the RNV problem. We also show how this problem can be used to achieve optimal query time for a number of interesting variants of the classic pattern matching problems. [Copyright &y& Elsevier]

Abstract: In this paper, we present algorithms that compute large matchings in planar graphs with fixed minimum degree. The algorithms give a guarantee on the size of the computed matching and run in linear time. Thus they are faster than the best known algorithm for computing maximum matchings in general graphs and in planar graphs, which run in and time, respectively. For the class of planar graphs with minimum degree 3, the bounds we achieve are known to be the best possible. Further, we discuss how minimum degree 5 can be used to obtain stronger bounds on the matching size. [Copyright &y& Elsevier]

This paper discusses the traditional philosophical debate concerning body and spirit that has formed a multidisciplinary field of researches about brain and mind in the twentieth century, called cognitive sciences. Among the various divergent positions in this area, three of them had their nuances about the future of this exploited enterprise, and also their relations with the social sciences that basically move in these directions: (i) the defence of the singularity of human sciences' object considered in the philosophical consciousness model as intentionality; (ii) the searching of a biological understanding of the mind, also viewed in this case as intentionality and consciousness, and finally; (iii) the critic of the notions of intentionality and consciousness through the cybernetic and also the cognitive sciences achievements, that can feed a new relation between humans and machines. [ABSTRACT FROM AUTHOR]

Abstract: This paper examines the computer science issues associated with secure remote instrumentation control for magnetic fusion experiments. Computer science research into enhancing the ability to scientifically participate in a fusion experiment remotely has been growing in size in an attempt to better address the needs of fusion scientists worldwide. The natural progression of this research is to examine how to move from remote scientific participation to remote hardware control. The vision is to define a gatekeeper software system that will be the only channel of interaction for incoming requests to the secured area of the experimental site. The role of the gatekeeper is to validate the identification and access privilege of the requestor and to insure the general validity of the proposed request. The vision for the gatekeeper is that it be a modular system that is simple in design and defined in a way that makes its implementation and operation transparent and obvious. The architecture of the module interface is flexible enough that it can easily allow the future addition of new modules. At the same time, it should be transparent to end-users and allow a high volume of activity so as to not provide a work bottleneck. The results of the gatekeeper design and initial implementation are presented as well as a discussion on the implication of this research on the operation of fusion experimental machines such as ITER. [ABSTRACT FROM AUTHOR]

Ant colony optimization (ACO) has widely been applied to solve combinatorial optimization problems in recent years. There are few studies, however, on its convergence time, which reflects how many iteration times ACO algorithms spend in converging to the optimal solution. Based on the absorbing Markov chain model, we analyze the ACO convergence time in this paper. First, we present a general result for the estimation of convergence time to reveal the relationship between convergence time and pheromone rate. This general result is then extended to a two-step analysis of the convergence time, which includes the following: 1) the iteration time that the pheromone rate spends on reaching the objective value and 2) the convergence time that is calculated with the objective pheromone rate in expectation. Furthermore, four brief ACO algorithms are investigated by using the proposed theoretical results as case studies. Finally, the conclusions of the case studies that the pheromone rate and its deviation determine the expected convergence time are numerically verified with the experiment results of four one-ant ACO algorithms and four tenant ACO algorithms. [ABSTRACT FROM AUTHOR]

This paper explores the architecture, theoretical foundations, and paradigms of contemporary cybernetics from perspectives of cognitive informatics (CI) and computational intelligence. The modern domain and the hierarchical behavioral model of cybernetics are elaborated at the imperative, autonomic, and cognitive layers. The CI facet of cybernetics is presented, which explains how the brain may be mimicked in cybernetics via CI and neural informatics. The computational intelligence facet is described with a generic intelligence model of cybernetics. The compatibility between natural and cybernetic intelligence is analyzed. A coherent framework of contemporary cybernetics is presented toward the development of transdisciplinary theories and applications in cybernetics, CI, and computational intelligence. [ABSTRACT FROM AUTHOR]

Abstract: One-way functions are a fundamental notion in cryptography, since they are the necessary condition for the existence of secure encryption schemes. Most examples of such functions, including Factoring, Discrete Logarithm or the RSA function, however, can be inverted with the help of a quantum computer. Hence, it is very important to study the possibility of quantum one-way functions, i.e. functions which are easily computable by a classical algorithm but are hard to invert even by a quantum adversary. In this paper, we provide a set of problems that are good candidates for quantum one-way functions. These problems include Graph Non-Isomorphism, Approximate Closest Lattice Vector and Group Non-Membership. More generally, we show that any hard instance of Circuit Quantum Sampling gives rise to a quantum one-way function. By the work of Aharonov and Ta-Shma [D. Aharonov, A. Ta-Shma, Adiabatic quantum state generation and statistical zero knowledge, in: Proceedings of STOC02 — Symposium on the Theory of Computing, 2001], this implies that any language in Statistical Zero Knowledge which is hard-on-average for quantum computers leads to a quantum one-way function. Moreover, extending the result of Impagliazzo and Luby [R. Impagliazzo, M. Luby, One-way functions are essential for complexity based cryptography, in: Proceedings of FOCS89 — Symposium on Foundations of Computer Science, 1989] to the quantum setting, we prove that quantum distributionally one-way functions are equivalent to quantum one-way functions. [Copyright &y& Elsevier]

Life sciences, and biology in particular, are heavily impacted by the development of methods for data collection and data analysis. Taking advantage of the availability of data, modeling biological effects is becoming more and more popular and relevant in life sciences, due to the diffuse and wide use of information technology (IT) tools. IT is increasing the availability of models, the horizon and complexity of modeling activities, reaching new targets, and boosting "virtuality". Modeling, as any inductive activity, originates from two needs: to predict future outcomes using previous experience, and to explain observations, or in other terms to infer knowledge. In this paper we analyze criteria, problems, possibilities and advancements which indicate that the time for e-modeling in biology and other life sciences is ripe. To do this, we take as example a particular field of biological sciences, the computational toxicity (CT) of chemicals, and its usual QSAR (Quantitative Structure Activity Relationships) approach. The open possibilities are evaluated, including a reshaping of the interface between toxicology, chemistry and computer science. The epistemological problem about "what models are" is approached in a pragmatic sense. [ABSTRACT FROM AUTHOR]

Abstract: This paper presents a modular framework for the specification of certain inductively-defined coalgebraic types. Modal logics for coalgebras of polynomial endofunctors on the category of sets have been studied in [M. Rößiger, Coalgebras and modal logic, in: H. Reichel (Ed.), Coalgebraic Methods in Computer Science, Electronic Notes in Theoretical Computer Science, vol. 33, Elsevier Science, 2000, pp. 299–320; B. Jacobs, Many-sorted coalgebraic modal logic: a model-theoretic study, Theoretical Informatics and Applications 35(1) (2001) 31–59]. These logics are here generalised to endofunctors on categories of sorted sets, in order to allow collections of inter-related types to be specified simultaneously. The inductive nature of the coalgebraic types considered is then used to formalise semantic relationships between different types, and to define translations between the associated logics. The resulting logical framework is shown to be an institution, whose specifications and specification morphisms admit final and respectively cofree models. [Copyright &y& Elsevier]

Abstract: In this paper, we present an out of order quantifier elimination algorithm for a class of Quantified Linear Programs (QLPs) called Standard Quantified Linear Programs (SQLPs). QLPs in general and SQLPs in particular are extremely useful constraint logic programming structures that find wide applicability in the modeling of real-time schedulability specifications; see Subramani [Subramani, K., 2005a. A comprehensive framework for specifying clairvoyance, constraints and periodicity in real-time scheduling. The Computer Journal 48 (3), 259–272]. Consequently any algorithmic advance in their solution has a strong practical impact. Prior to this work, the only known approaches to the solution of QLPs involved sequential variable elimination; see Subramani [Subramani, K., 2003b. An analysis of quantified linear programs. In: Calude, C.S. et al. (Eds.), Proceedings of the 4th International Conference on Discrete Mathematics and Theoretical Computer Science. DMTCS. In: Lecture Notes in Computer Science, vol. 2731. Springer-Verlag, pp. 265–277]. In the sequential approach, the innermost quantified variable is eliminated first, followed by the variable which then becomes the innermost quantified variable and so on, until we are left with a single variable from which the satisfiability of the original formula is easily deduced. This approach is applicable in both discrete and continuous domains; however, it is to be noted that the logic demanding the sequential approach requires that the variables are discrete-valued. To the best of our knowledge, the necessity for sequential elimination over continuous-valued variables has not been investigated in the literature. The techniques used in the development of our elimination algorithm may find applications in domains such as classical logic and finite model theory. The final aspect of our research concerns the structure-preserving nature of the algorithm that we introduce here; in general, it is not known whether discrete domains admit such elimination procedures. [Copyright &y& Elsevier]

Abstract: More than a dozen years ago, Amadio [Bifinite domains: stable case, in: Lecture Notes in Computer Science, vol. 530, 1991, pp. 16–33] (see Amadio and Curien, Domains and Lambda-Calculi, Cambridge Tracts in Theoretical Computer Science, vol. 46, Cambridge University Press, 1998 as well) raised the question of whether the category of stable bifinite domains of Amadio–Droste [R.M. Amadio, Bifinite domains: stable case, in: Lecture Notes in Computer Science, vol. 530, 1991, pp. 16–33; M. Droste, On stable domains, Theor. Comput. Sci. 111 (1993) 89–101; M. Droste, Cartesian closed categories of stable domains for polymorphism, Preprint, Universität GHS Essen] is the largest cartesian closed full sub-category of the category of ω-algebraic meet-cpos with stable functions. An affirmative solution to this problem has two major steps: (1) Show that for any ω-algebraic meet-cpo D, if all higher-order stable function spaces built from D are ω-algebraic, then D is finitary (i.e., it satisfies the so-called axiom I); (2) Show that for any ω-algebraic meet-cpo D, if D violates MI ∞, then [D→D] violates either M or I. We solve the first part of the problem in this paper, i.e., for any ω-algebraic meet-cpo D, if the stable function space [D→D] satisfies M, then D is finitary. Our notion of (mub, meet)-closed set, which is introduced for step 1, will also be used for treating some example cases in step 2. [Copyright &y& Elsevier]

Abstract: Many physical events assume values that can be represented with functions that are monotonic with respect to time passing. This paper defines the class of Monotonic Hybrid Systems, namely Hybrid Systems where activities and assignments are monotonic functions. On the basis that, among Hybrid Systems, Integrator Systems (which are Hybrid Systems equipped with stopwatches) are the most expressive of the classes with a linear evolution law, and that Timed Systems (which are Hybrid Systems equipped with clocks) enjoy most of the decidable properties required, we compare Monotonic Hybrid Systems with the above classes, with respect to decidability, expressiveness and succinctness. We show that a subclass of Monotonic Hybrid Systems is equivalent to Integrator Systems and Timed Systems with discrete time assumption, but that it strictly contains them when assuming dense time. This has practical consequences both in the discrete and dense context when symbolic verification is not possible. We also show that Monotonic Hybrid Systems allow more succinct descriptions than those of Integrator and Timed Systems. This result shows that not only hierarchy, non-determinism and communication, as already noticed in Alur et al. (Proceedings of the HSCC 00, Lecture Notes in Computer Science, Springer, Berlin, 2000, pp. 6–19) and Grosu and Stauner (Formal Methods System Design) 21 (2002) 5), have an impact on succinctness, but also the shape of functions used in descriptions does. For a subclass of Monotonic Hybrid Systems for which reachability is decidable if the functions used are computable, we consider both the problem of verifying properties written in a suitable logic and also the relationship between dense time and discrete time assumptions in the framework of verification. The formalism and the results are illustrated by some examples. [Copyright &y& Elsevier]

A comparison between the development of cognitive science in France and the USA enables us to analyze some national differences linked to specific connections between the scientific, military, economic and political worlds. The influence of new practices and tools developed during World War II and the Cold War appears to be of crucial importance in understanding the development of this new field, as well as that of cybernetics, computer science, artificial intelligence and molecular biology. This paper can be considered as a study in how the differing contexts in France and the USA shaped the history of the construction of cognitive science in each of these two countries. In spite of various differences, some common aspects may be pointed out: in both cases, computer experts and psychologists using a computational modelling approach were those first engaged in the construction of cognitive science. If in France neuroscience-oriented cognitive science research was stronger than in the USA, it seems that the artificial intelligence orientation is also of growing importance in France. [ABSTRACT FROM AUTHOR]

The discipline of software engineering has transferred the commonsense methods of good programming and management to large software projects, it has been less successful in acquiring a solid theoretical foundation for these methods. We have developed an introductory computer science course, much as calculus is a basic course for mathematics and the physical sciences, concerned primarily with theoretical foundations and methodology rather than apprenticeship through applications. This paper describes the principles taught in the course and gives a small example illustrating them. [ABSTRACT FROM AUTHOR]

In the above paper [1], acknowledgment for financial support is missing. The statement is provided in the footnote below. [ABSTRACT FROM PUBLISHER]

This section is maintained by Vladik Kreinovich. Please send your abstracts (or copies of papers that you want to see reviewed here) to vladik@cs.utep.edu, or by regular mail to: VLADIK KREINOVICH, Department of Computer Science, University of Texas at El Paso, El Paso, TX 79968, USA. [ABSTRACT FROM AUTHOR]

Tomaso Poggio is Eugene McDermott Professor in the Department of Brain and Cognitive Sciences at MIT, where he is also member of the Computer Science and Artificial Intelligence Lab and of the new McGovern Institute for Brain Research. He received his doctorate in physics in Italy, then worked for ten years in the Max Planck Institute for Biological Cybernetics in Tübingen, Germany, before moving to MIT where he has been for the last 23 years. His papers span mathematics, computer science and neuroscience. His main contributions are in computer and biological vision (in both flies and humans), computational neuroscience, system theory and computer graphics. He currently leads a group of researchers investigating the problem of learning in terms of the underlying mathematics, its engineering applications and its role for understanding object recognition in visual cortex. [Copyright &y& Elsevier]