INDUCTIVE LEARNING FOR PARAMETER OPTIMIZATION

Optimization of simulation model output is one of the most important tasks in a simulation study of a complex system. Efficacy of an optimization approach is expressed in the accuracy of locating a global extremum, as well as in the number of investigated search points. The approach Machine Learning Optimization (ML-Opt), presented in this article, explores functional dependencies between search points in order to reduce the number of evaluations. Functional relations between search points are determined by an inductive learning algorithm, which generates a classifier used as a control structure in the optimization process. The classifier approximates the structure of the unknown goal function given by a simulation model and affects the generation of new search points. A discussion of a numerical example concludes the paper.     

Beiträge des WBW zur Entwicklung der Logistik in Forschung und Lehre

On the occasion of the 50th anniversary, the contributions of the Department of Economic and Management Sciences (WBW), University of Leoben, to the development of logistics in research and education are highlighted. After a discussion of the milestones of logistics and related dissertations of the department WBW, the newly resulting curriculum is presented. The paper concludes with an outlook on a possible new research topic.     

Paying to belong: When does rejection trigger ingratiation?

Societies and social scientists have long held the belief that exclusion induces ingratiation and conformity, an idea in contradiction to robust empirical evidence linking rejection with hostility and aggression. The classic literatures on ingratiation and conformity help resolve this contradiction by identifying circumstances under which rejection may trigger efforts to ingratiate. Jointly, findings from these literatures suggest that when people are given an opportunity to impress their rejecters, ingratiation is likely after rejection experiences that are harsh and that occur in important situations that threaten the individual's self-definition. Four studies tested the hypothesis that people high in rejection sensitivity and therefore dispositionally concerned about rejection will utilize opportunities to ingratiate after harsh rejection in situations that are self-defining. In 3 studies of situations that are particularly self-defining for men, rejection predicted ingratiation among men (but not women) who were high in rejection sensitivity. In a 4th study, harsh rejection in a situation particularly self-defining for women predicted ingratiation among highly rejection-sensitive women (but not men). These findings help identify the specific circumstances under which people are willing to act in socially desirable ways toward those who have rejected them harshly. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The Iterated Weakest Link

We outline a model for security investment that reflects dynamic interaction between a defender, who faces uncertainty, and an attacker, who repeatedly targets the weakest link. Using the model, we derive and compare optimal security investment over multiple periods, exploring the delicate balance between proactive and reactive security investment.     

Drying quality of rubberwood: an industrial perspective

In this study, the kiln drying yield of 30 mm Rubberwood (Hevea brasiliensis) used for the furniture manufacturing industry in South East Asia was evaluated. The results showed that the average yield loss was 10%, and end-checks, splits and warping were the most common drying defects. It was found that sawn timber from older trees (> 21 years) were less prone to drying defects, and the use of stress relieving treatments significantly reduced the incidence of drying defects.     

Inverse 1-center location problems with edge length augmentation on trees

This paper considers the inverse 1-center location problem with edge length augmentation on a tree network T with n + 1 vertices. The goal is to increase the edge lengths at minimum total cost subject to given modification bounds such that a predetermined vertex s becomes an absolute 1-center under the new edge lengths. Using a set of suitably extended AVL-search trees we develop a combinatorial algorithm which solves the inverse 1-center location problem with edge length augmentation in O(n log n) time. Moreover, it is shown that the problem can be solved in O(n) time if all the cost coefficients are equal.     

Methodological issues in the validation of implicit measures: Comment on De Houwer, Teige-Mocigemba, Spruyt, and Moors (2009).

J. De Houwer, S. Teige-Mocigemba, A. Spruyt, and A. Moors’s (see record 2009-05290-001) normative analysis of implicit measures provides an excellent clarification of several conceptual ambiguities surrounding the validation and use of implicit measures. The current comment discusses an important, yet unacknowledged, implication of J. De Houwer et al.’s analysis, namely, that investigations addressing the proposed implicitness criterion (i.e., does the relevant psychological attribute influence measurement outcomes in an automatic fashion?) will be susceptible to fundamental misinterpretations if they are conducted independently of the proposed what criterion (i.e., is the measurement outcome causally produced by the psychological attribute the measurement procedure was designed to assess?). As a solution, it is proposed that experimental validation studies should be combined with a correlational approach in order to determine whether a given manipulation influenced measurement scores via variations in the relevant psychological attribute or via secondary sources of systematic variance. (PsycINFO Database Record (c) 2010 APA, all rights reserved)