The double-edged effects of explanation prompts

Explanation prompts usually foster conceptual understanding. However, it has been claimed within cognitive load theory that prompts can take cognitive load to the upper limit when learning complex contents. Under such circumstances, prompts focusing the learners’ attention on specific aspects (e.g., conceptual aspects such as elaborations on domain principles) might have some costs: Other important aspects (e.g., procedural aspects such as how to calculate) cannot be processed deeply. Thus, we expected that conceptually-oriented explanation prompts would foster the detailedness of explanations, the number of elaborations on domain principles, and conceptual knowledge. In addition, we tested the influence of such prompts on the number of calculations performed during learning and procedural knowledge. We conducted an experiment in which we employed conceptually-oriented explanation prompts in a complex e-learning module on tax law. Tax law university students (N = 40) worked on this e-learning module under two conditions: (a) conceptually-oriented explanation prompts, (b) no prompts. The prompts led to double-edged effects: positive effects on the detailedness of explanations and on the number of elaborations on domain principles, as well as on conceptual knowledge and simultaneously negative effects on the number of calculations performed during learning as well as on procedural knowledge.     

To Score or Not to Score? Tripling Insights for Participatory Design

For evaluating visual-analytics tools, many studies confine to scoring user insights into data. For participatory design of those tools, we propose a three-level methodology to make more out of users' insights. The Relational Insight Organizer (RIO) helps to understand how insights emerge and build on one each other.     

New global optimization algorithms for model-based clustering

The Expectation-Maximization (EM) algorithm is a very popular optimization tool for mixture problems and in particular for model-based clustering problems. However, while the algorithm is convenient to implement and numerically very stable, it only produces local solutions. Thus, it may not achieve the globally optimal solution in problems that have a large number of local optima. This paper introduces several new algorithms designed to produce global solutions in model-based clustering. The building blocks for these algorithms are methods from the operations research literature, namely the Cross-Entropy (CE) method and Model Reference Adaptive Search (MRAS). One problem with applying these methods directly is the efficient simulation of positive definite covariance matrices. We propose several new solutions to this problem. One solution is to apply the principles of Expectation-Maximization updating, which leads to two new algorithms, CE-EM and MRAS-EM. We also propose two additional algorithms, CE-CD and MRAS-CD, which rely on the Cholesky decomposition. We conduct numerical experiments of varying complexity to evaluate the effectiveness of the proposed algorithms in comparison to classical EM. We find that although a single run of the new algorithms is slower than a single run of EM, all have the potential for producing significantly better solutions. We also find that although repeat application of EM may achieve similar results, our algorithms provide automated, data-driven decision rules which may significantly reduce the burden of searching for the global optimum.     

Unsupervised strategies for shilling detection and robust collaborative filtering

Collaborative filtering systems are essentially social systems which base their recommendation on the judgment of a large number of people. However, like other social systems, they are also vulnerable to manipulation by malicious social elements. Lies and Propaganda may be spread by a malicious user who may have an interest in promoting an item, or downplaying the popularity of another one. By doing this systematically, with either multiple identities, or by involving more people, malicious user votes and profiles can be injected into a collaborative recommender system. This can significantly affect the robustness of a system or algorithm, as has been studied in previous work. While current detection algorithms are able to use certain characteristics of shilling profiles to detect them, they suffer from low precision, and require a large amount of training data. In this work, we provide an in-depth analysis of shilling profiles and describe new approaches to detect malicious collaborative filtering profiles. In particular, we exploit the similarity structure in shilling user profiles to separate them from normal user profiles using unsupervised dimensionality reduction. We present two detection algorithms; one based on PCA, while the other uses PLSA. Experimental results show a much improved detection precision over existing methods without the usage of additional training time required for supervised approaches. Finally, we present a novel and highly effective robust collaborative filtering algorithm which uses ideas presented in the detection algorithms using principal component analysis.     

Changes in the microstructure of materials and their impact on reliability: experiments and modeling

Several types of microstructural changes in microelectronic solder materials are briefly discussed, namely the formation of intermetallics, Kirkendall voiding, spinodal decomposition, and coarsening. Moreover, a theoretical framework, based on the entropy principle, is developed by means of which constitutive equations can be derived that are necessary to understand and simulate such processes. The resulting equations are specialized to binary alloys and then investigated numerically. Finally, simulations of spinodal decomposition and coarsening are performed for the eutectic solder AgCu. The theoretical predictions are compared with experiments.

---

Pupil detection for head-mounted eye tracking in the wild: an evaluation of the state of the art

Robust and accurate detection of the pupil position is a key building block for head-mounted eye tracking and prerequisite for applications on top, such as gaze-based human–computer interaction or attention analysis. Despite a large body of work, detecting the pupil in images recorded under real-world conditions is challenging given significant variability in the eye appearance (e.g., illumination, reflections, occlusions, etc.), individual differences in eye physiology, as well as other sources of noise, such as contact lenses or make-up. In this paper we review six state-of-the-art pupil detection methods, namely ElSe (Fuhl et al. in Proceedings of the ninth biennial ACM symposium on eye tracking research & applications, ACM. New York, NY, USA, pp 123–130, 2016), ExCuSe (Fuhl et al. in Computer analysis of images and patterns. Springer, New York, pp 39–51, 2015), Pupil Labs (Kassner et al. in Adjunct proceedings of the 2014 ACM international joint conference on pervasive and ubiquitous computing (UbiComp), pp 1151–1160, 2014. doi: 10.1145/2638728.2641695), SET (Javadi et al. in Front Neuroeng 8, 2015), Starburst (Li et al. in Computer vision and pattern recognition-workshops, 2005. IEEE Computer society conference on CVPR workshops. IEEE, pp 79–79, 2005), and ?wirski (?wirski et al. in Proceedings of the symposium on eye tracking research and applications (ETRA). ACM, pp 173–176, 2012. doi: 10.1145/2168556.2168585). We compare their performance on a large-scale data set consisting of 225,569 annotated eye images taken from four publicly available data sets. Our experimental results show that the algorithm ElSe (Fuhl et al. 2016) outperforms other pupil detection methods by a large margin, offering thus robust and accurate pupil positions on challenging everyday eye images.     

A complete strategy for efficient and accurate multibody dynamics of flexible structures with large lap joints considering contact and friction

This paper deals with the dynamics of jointed flexible structures in multibody simulations. Joints are areas where the surfaces of substructures come into contact, for example, screwed or bolted joints. Depending on the spatial distribution of the joint, the overall dynamic behavior can be influenced significantly. Therefore, it is essential to consider the nonlinear contact and friction phenomena over the entire joint. In multibody dynamics, flexible bodies are often treated by the use of reduction methods, such as component mode synthesis (CMS). For jointed flexible structures, it is important to accurately compute the local deformations inside the joint in order to get a realistic representation of the nonlinear contact and friction forces. CMS alone is not suitable for the capture of these local nonlinearities and therefore is extended in this paper with problem-oriented trial vectors. The computation of these trial vectors is based on trial vector derivatives of the CMS reduction base. This paper describes the application of this extended reduction method to general multibody systems, under consideration of the contact and friction forces in the vector of generalized forces and the Jacobian. To ensure accuracy and numerical efficiency, different contact and friction models are investigated and evaluated. The complete strategy is applied to a multibody system containing a multilayered flexible structure. The numerical results confirm that the method leads to accurate results with low computational effort.     

A scaled thickness conditioning for solid- and solid-shell discretizations of thin-walled structures

The numerical discretization of thin shell structures yields ill-conditioned stiffness matrices due to an inherent large eigenvalue spectrum. Finite element parametrization that depends on shell thickness, like relative displacement shells, solid shells and other solid finite elements even add to the ill-conditioning by introducing high eigenmodes.To overcome this numerical issue we present a scaled thickness conditioning (STC) approach, a mechanically motivated preconditioner for thin-walled structures discretized with continuum based element formulations. The proposed approach is motivated by the scaled director conditioning (SDC) method for relative displacement shell elements. In contrast to SDC, the novel STC approach yields a preconditioner for the effective linear system. It is applicable independently of element technology employed, coupling to other physical fields, boundary conditions applied and additional algebraic constraints and can be easily extended to multilayer shell formulations.The effect of the proposed preconditioner on the conditioning of the effective stiffness matrix and its eigenvalue spectrum is studied. It is shown that the condition number of the modified system becomes almost independent from the aspect ratio of the employed elements. The improved conditioning has a positive influence on the convergence behavior of iterative linear solvers. In particular, in combination with algebraic multigrid preconditioners the number of iterations could be decreased by more than 85% for some examples and the computation time could be reduced by about 60%.     

Isochronous sets of invariant control systems

Let G be a connected Lie group with Lie algebra g and Σ=(G,D) a controllable invariant control system. A subset A⊂G is said to be isochronous if there exists a uniform time T_A>0 such that any two arbitrary elements in A can be connected by a positive orbit of Σ at exact time T_A. In this paper, we search for classes of Lie groups G such that any Σ has the following property: there exists an increasing sequence of open neighborhoods (V_n)_n≥0 of the identity in G such that the group can be decomposed in isochronous rings W_n=V_n+1−V_n. We characterize this property in algebraic terms and we show that three classes of Lie groups satisfy this property: completely solvable simply connected Lie groups, semisimple Lie groups and reductive Lie groups.     

Evolving Teams of Predictors with Linear Genetic Programming

This paper applies the evolution of GP teams to different classification and regression problems and compares different methods for combining the outputs of the team programs. These include hybrid approaches where (1) a neural network is used to optimize the weights of programs in a team for a common decision and (2) a realnumbered vector (the representation of evolution strategies) of weights is evolved with each term in parallel. The cooperative team approach results in an improved training and generalization performance compared to the standard GP method. The higher computational overhead of team evolution is counteracted by using a fast variant of linear GP.