Connecting Community-Grids by supporting job negotiation with coevolutionary Fuzzy-Systems

We utilize a competitive coevolutionary algorithm (CA) in order to optimize the parameter set of a Fuzzy-System for job negotiation between Community-Grids. In a Community-Grid, users are submitting jobs to their local High Performance Computing (HPC) sites over time. Now, we assume that Community-Grids are interconnected such that the exchange of jobs becomes possible: Each Community strives for minimizing the response time for their own members by trying to distribute workload to other communities in the Grid environment. For negotiation purpose, a Fuzzy-System is used to steer each site’s decisions whether to distribute or accept workload in a beneficial, yet egoistic direction. In such a system, it is essential that communities can only benefit if the workload is equitably (not necessarily equally) portioned among all participants. That is, if one community egoistically refuses to execute foreign jobs regularly, other HPC sites suffer from overloading. This, on the long run, deteriorates the opportunity to utilize them for job delegation. Thus, the egoistic community will degrade its own average performance. This scenario is particularly suited for the application of a competitive CA: the Fuzzy-Systems of the participating communities are modeled as species, which evolve in different populations while having to compete within the commonly shared ecosystem. Using real workload traces and Grid setups, we show that the opportunistic cooperation leads to significant improvements for both each community and the overall system.     

Counting decomposable multivariate polynomials

A polynomial f (multivariate over a field) is decomposable if f=g °h{f=g \circ h} with g univariate of degree at least 2. We determine the dimension (over an algebraically closed field) of the set of decomposables, and an approximation to their number over a finite field. The relative error in our approximations is exponentially decaying in the input size.     

Learning with few examples for binary and multiclass classification using regularization of randomized trees

The human visual system is often able to learn to recognize difficult object categories from only a single view, whereas automatic object recognition with few training examples is still a challenging task. This is mainly due to the human ability to transfer knowledge from related classes. Therefore, an extension to Randomized Decision Trees is introduced for learning with very few examples by exploiting interclass relationships. The approach consists of a maximum a posteriori estimation of classifier parameters using a prior distribution learned from similar object categories. Experiments on binary and multiclass classification tasks show significant performance gains     

The goal specificity effect on strategy use and instructional efficiency during computer-based scientific discovery learning

Using a computer-based scientific discovery learning environment on buoyancy in fluids we investigated the effects of goal specificity (nonspecific goals vs. specific goals) for two goal types (problem solving goals vs. learning goals) on strategy use and instructional efficiency. Our empirical findings close an important research gap, because in earlier studies the goal specificity effect either was restricted to one goal type or goal type was confounded with goal specificity. In addition, there is hardly a study with empirical evidence for the goal specificity effect on strategy use, which counts even more for a cognitive cost-benefit ratio as a dependent variable. Instead, in earlier studies the goal specificity effect has been attributed to differences in strategy use and cognitive cost-benefit ratio in a rather theoretical way. In the present study for strategy use an interaction was found between goal specificity and goal type, indicating that the goal specificity effect occurs only in case of problem solving goals, but not in case of learning goals. Compared to students provided with specific problem solving goals, students who worked on nonspecific problem solving goals, used a control of variables-strategy more frequently. Additionally, we found a main effect of goal specificity on instructional efficiency for both of the goal types, pointing at a more favorable relationship between performance gain and cognitive load caused by nonspecific goals.     

Can Variational Models for Correspondence Problems Benefit from Upwind Discretisations?

Optic flow and stereo reconstruction are important examples of correspondence problems in computer vision. Correspondence problems have been studied for almost 30 years, and energy-based methods such as variational approaches have become popular for solving this task. However, despite the long history of research in this field, only little attention has been paid to the numerical approximation of derivatives that naturally occur in variational approaches.     

Organized Nanostructured Complexes of Polyoxometalates and Surfactants that Exhibit Photoluminescence and Electrochromism

A variety of functional nanostructured organic/inorganic hybrid materials from the europium‐exchanged derivative of a Preyssler‐type polyoxometalate (POM), [EuP₅W₃₀O₁₁₀]^12?, and functional organic surfactants were prepared by the ionic self‐assembly (ISA) route. The effect of organic surfactants on the structure, photoluminescent, electrochemical and electrochromic properties of the POM anions was investigated in detail. All obtained hybrid materials are amphotropic, i.e., exhibit both thermotropic and lyotropic liquid‐crystalline phase behaviour. Investigations of their photophysical properties have shown that the interactions of the various surfactants with the polyanions influence the coordination environments and site symmetry of Eu³⁺ in different ways. The functional groups in the organic surfactants significantly influence the electrochromic properties and photoluminescence of POMs. Different from normal and pyridine‐containing complexes, no photoluminescence and no electrochromism were observed from the ferrocene‐containing complexes. This may be explained in view of charge transfer between the POM anion and the ferrocenyl group.     

Playing chess unconsciously.

Expertise in a certain stimulus domain enhances perceptual capabilities. In the present article, the authors investigate whether expertise improves perceptual processing to an extent that allows complex visual stimuli to bias behavior unconsciously. Expert chess players judged whether a target chess configuration entailed a checking configuration. These displays were preceded by masked prime configurations that either represented a checking or a nonchecking configuration. Chess experts, but not novice chess players, revealed a subliminal response priming effect, that is, faster responding when prime and target displays were congruent (both checking or both nonchecking) rather than incongruent. Priming generalized to displays that were not used as targets, ruling out simple repetition priming effects. Thus, chess experts were able to judge unconsciously presented chess configurations as checking or nonchecking. A 2nd experiment demonstrated that experts' priming does not occur for simpler but uncommon chess configurations. The authors conclude that long-term practice prompts the acquisition of visual memories of chess configurations with integrated form-location conjunctions. These perceptual chunks enable complex visual processing outside of conscious awareness. (PsycINFO Database Record (c) 2010 APA, all rights reserved)