Content uploaded by Mikael Collan
Author content
All content in this area was uploaded by Mikael Collan on Sep 09, 2016
Content may be subject to copyright.
A preview of the PDF is not available
Proceedings of NSAIS16 Workshop on Intelligent and Adaptive Systems
Abstract and Figures
This workshop NSAIS´16 is the second one in the series of workshops organized by the
North-European Society for Adaptive and Intelligent Systems (NSAIS). The first workshop
took place more than ten years ago and this is the first international event organized by
NSAIS ever since. The idea of this workshop is to be the beginning of a series of workshops
that take place typically bi-annually around the areas of interest to the NSAIS in Northern
Europe.
This workshop is organized in association with the Lappeenranta University of Technology
(LUT), the Finnish Operations Research Society (FORS), TRIZ Finland HUB, and the
European Society for Fuzzy Logic and Technology (EUSFLAT) – the organizers thank the
aforementioned organizations for their support.
This proceedings includes altogether seventeen papers, short papers, or abstracts of the
submissions presented at the NSAIS´16 workshop that represent nine nationalities. The papers
are a mix along the topical theme of the conference “OR+Fuzzy+TRIZ” with many
contributions to decision-making. All papers have undergone peer review. The organizers
thank the international scientific program committee and the reviewers for their support.
Figures - uploaded by Mikael Collan
Author content
All figure content in this area was uploaded by Mikael Collan
Content may be subject to copyright.
ResearchGate has not been able to resolve any citations for this publication.
Too many new products fail. New products which are hard to differentiate from existing products won’t capture the customer’s imagination. The failure is due to a poor understanding of customers’ needs. Companies need to take a radical approach to identifying customers’ real needs, and this book demonstrates innovative ways to achieve this.
This article introduces a multi-criteria decision making (MCDM) method that takes into account interdependency of criteria. Recognizing information redundancy in related criteria the proposed approach offers their weight formation based on their interaction. Further, normalized estimated criteria and their weights are aggregated by means of Fuzzy Heavy Weighted Averaging (FHWA) operator. The approach is illustrated with a numerical case study of engineering design selection problem. Keywords— Multi-criteria decision making, design assessment, fuzzy logic, information redunduncy
The methodology used to construct tree structured rules is the focus of this monograph. Unlike many other statistical procedures, which moved from pencil and paper to calculators, this text's use of trees was unthinkable before computers. Both the practical and theoretical sides have been developed in the authors' study of tree methods. Classification and Regression Trees reflects these two sides, covering the use of trees as a data analysis method, and in a more mathematical framework, proving some of their fundamental properties.
p>A supervised learning algorithm (Scaled Conjugate Gradient, SCG) with superlinear convergence rate is introduced. The algorithm is based upon a class of optimization techniques well known in numerical analysis as the Conjugate Gradient Methods. SCG uses second order information from the neural network but requires only O(N) memory usage, where N is the number of weights in the network. The performance of SCG is benchmarked against the performance of the standard backpropagation algorithm (BP), the conjugate gradient backpropagation (CGB) and the one-step Broyden-Fletcher-Goldfarb-Shanno memoryless quasi-Newton algorithm (BFGS). SCG yields a speed-up of at least an order of magnitude relative to BP. The speed-up depends on the convergence criterion, i.e., the bigger demand for reduction in error the bigger the speed-up. SCG is fully automated including no user dependent parameters and avoids a time consuming line-search, which CGB and BFGS use in each iteration in order to determine an appropriate step size.
Incorporating problem dependent structural information in the architecture of a neural network often lowers the overall complexity. The smaller the complexity of the neural network relative to the problem domain, the bigger the possibility that the weight space contains long ravines characterized by sharp curvature. While BP is inefficient on these ravine phenomena, it is shown that SCG handles them effectively.</p
Thesupport-vector network is a new learning machine for two-group classification problems. The machine conceptually implements the following idea: input vectors are non-linearly mapped to a very high-dimension feature space. In this feature space a linear decision surface is constructed. Special properties of the decision surface ensures high generalization ability of the learning machine. The idea behind the support-vector network was previously implemented for the restricted case where the training data can be separated without errors. We here extend this result to non-separable training data.High generalization ability of support-vector networks utilizing polynomial input transformations is demonstrated. We also compare the performance of the support-vector network to various classical learning algorithms that all took part in a benchmark study of Optical Character Recognition.
Uncertainty about the future course of the economy is a potential driver of aggregate fluctuations. To identify the distinct dimensions of uncertainty in the macroeconomy, we construct a large dataset covering all types of economic uncertainty. We then identify two fundamental factors that account for the common dynamics in this dataset. These factors are interpreted as macroeconomic uncertainty. The first factor captures business cycle uncertainty, while the second factor represents oil and commodity price uncertainty. While both types of uncertainty generate a decline in output, time-varying oil and commodity price uncertainty is more important for fluctuations in real activity. However, nonlinearities seem to amplify the effect of business cycle uncertainty during the global financial crisis. (JEL C32, C38, E32)