Suchergebnisse

Simulated annealing is a general approach for approximately solving large combinatorial optimization problems. The algorithm is based on an intriguing combination of ideas from at first sight completely unrelated fields of science, viz. combinatorial optimization and statistical physics. On the one hand the algorithm can be viewed as an analogue of an algorithm used in statistical physics for computer simulation of the annealing of a solid to its minimum–energy state, on the other hand it can be considered as a generalization of the well–known iterative improvement approach to combinatorial optimization problems.In this introductory paper we give a mathematical description of the simulated annealing algorithm and discuss its behaviour from both a theoretical and a practical point of view. The latter is illustrated by applying the algorithm to the travelling salesman problem.This paper was written to familiarize the readers of Statistica Neerlandica with simulated annealing. It is a summary of papers written earlier by the authors and does not contain any new material.

The work of Crocco and her colleagues, "Deliberating Public Policy Issues with Adolescents," combines two important fields—deliberative democracy and discussion as a pedagogy—with a study of policy deliberations in three classrooms. Their article yields valuable insights. As the authors note, the results are disappointing. This may be because the students were not actually asked to deliberate, if "deliberation" means discussing in order to make a decision. After all, the students could not decide US policy on immigration. Their discussion was a kind of simulated deliberation. Evidence suggests that we may see better results from real deliberations that occur within student-led voluntary associations or from simulated deliberations in which the students role-play powerful decision-makers, rather than playing themselves in a discussion that has no political impact.

Altres ajuts: Government of Spain/FEDER PGC2018-094364-B-I00 ; This paper studies method of simulated moments (MSM) estimators that are implemented using Bayesian methods, specifically Markov chain Monte Carlo (MCMC). Motivation and theory for the methods is provided by Chernozhukov and Hong (2003). The paper shows, experimentally, that confidence intervals using these methods may have coverage which is far from the nominal level, a result which has parallels in the literature that studies overidentified GMM estimators. A neural network may be used to reduce the dimension of an initial set of moments to the minimum number that maintains identification, as in Creel (2017). When MSM-MCMC estimation and inference is based on such moments, and using a continuously updating criteria function, confidence intervals have statistically correct coverage in all cases studied. The methods are illustrated by application to several test models, including a small DSGE model, and to a jump-diffusion model for returns of the S&P 500 index.

We are developing a practical-scale mechanical decladder that can slit nuclear spent fuel rod-cuts (hulls + pellets) on the order of several tens of kgf of heavy metal/batch to supply UO2 pellets to a voloxidation process. The mechanical decladder is used for separating and recovering nuclear fuel material from the cladding tube by horizontally slitting the cladding tube of a fuel rod. The Korea Atomic Energy Research Institute (KAERI) is improving the performance of the mechanical decladder to increase the recovery rate of pellets from spent fuel rods. However, because actual nuclear spent fuel is dangerously toxic, we need to develop simulated spent fuel rods for continuous experiments with mechanical decladders. We describe procedures to develop both simulated cladding tubes and simulated fuel rod (with physical properties similar to those of spent nuclear fuel). Performance tests were carried out to evaluate the decladding ability of the mechanical decladder using two types of simulated fuel (simulated tube + brass pellets and zircaloy-4 tube + simulated ceramic fuel rod). The simulated tube was developed for analyzing the slitting characteristics of the cross section of the spent fuel cladding tube. Simulated ceramic fuel rod (with mechanical properties similar to the pellets of actual PWR spent fuel) was produced to ensure that the mechanical decladder could slit real PWR spent fuel. We used castable powder pellets that simulate the compressive stress of the real spent UO2 pellet. The production criteria for simulated pellets with compressive stresses similar to those of actual spent fuel were determined, and the castables were inserted into zircaloy-4 tubes and sintered to produce the simulated fuel rod. To investigate the slitting characteristics of the simulated ceramic fuel rod, a verification experiment was performed using a mechanical decladder.

A stochastic model is proposed to describe the perceptual processes that underlie the classification of complex multidimensional sounds, such as passive sonar signatures. The model is based on four assumptions: (1) complex sounds may be represented psychologically as points in a multidimensional perceptual space, (2) categories are represented theoretically as distributions in this space centered at ideal category members or prototype, (3) the category distributions determine the conditional stimulus probabilities and have multivariate Gaussian form, and (4) individual dimensions in the perceptual space may be weighted differentially to reflect a listener's attentional processes. To evaluate the model, two groups were tested in an eight-category classification task involving 16 simulated propeller cavitation sounds. A perceptual structure was determined for the sounds in a preliminary non-metric multidimensional scaling study. The two groups received category assignments that stressed different properties of the cavitation sounds. An observed confusion matrix was derived by fitting the model to the observed data. Overall, comparison of the obtained and theoretical data indicated that the model provides a reasonable description of how listeners classify complex simulated sonar sounds. The assumption that listeners can selectively and independently adjust the relative importance of the perceptual dimensions was also supported. The utility of the proposed model as a tool for the development of preprocessing aids and training programs for sonar operators is discussed.

As policy makers require more rigorous assessments for the strength of evidence in Theory-Based evaluations, Bayesian logic is attracting increasing interest; however, the estimation of probabilities that this logic (almost) inevitably requires presents challenges. Probabilities can be estimated on the basis of empirical frequencies, but such data are often unavailable for most mechanisms that are objects of evaluation. Subjective probability elicitation techniques are well established in other fields and potentially applicable, but they present potential challenges and might not always be feasible. We introduce the community to a third way: simulated probabilities. We provide proof of concept that simulation can be used to estimate probabilities in diagnostic evaluation and illustrate our case with an application to health policy.

Objective: This research aims to identify the impact of text messaging on simulated driving performance. Background: In the past decade, a number of on-road, epidemiological, and simulator-based studies reported the negative impact of talking on a cell phone on driving behavior. However, the impact of text messaging on simulated driving performance is still not fully understood. Method: Forty participants engaged in both a single task (driving) and a dual task (driving and text messaging) in a high-fidelity driving simulator. Results: Analysis of driving performance revealed that participants in the dual-task condition responded more slowly to the onset of braking lights and showed impairments in forward and lateral control compared with a driving-only condition. Moreover, text-messaging drivers were involved in more crashes than drivers not engaged in text messaging. Conclusion: Text messaging while driving has a negative impact on simulated driving performance. This negative impact appears to exceed the impact of conversing on a cell phone while driving. Application: The results increase our understanding of driver distraction and have potential implications for public safety and device development.