A multi-objective approach to RBF network learning

The problem of inductive supervised learning is discussed in this paper within the context of multi-objective (MOBJ) optimization. The smoothness-based apparent (effective) complexity measure for RBF networks is considered. For the specific case of RBF network, bounds on the complexity measure are formally described. As the synthetic and real-world data experiments show, the proposed MOBJ learning method is capable of efficient generalization control along with network size reduction.     

On the impact of a more realistic physical layer on MANET simulations results

Network simulation tools are frequently used to analyze the performance of MANET protocols and applications. Currently they offer only simple wireless communication models that neglect many radio propagation effects. In this paper, we integrate a more realistic physical layer model into a simulation tool. It consists of a radio propagation model and a model of wireless transmission errors. They have been evaluated against real-world measurements. We prove that such more realistic models change simulation results considerably. To the best of our knowledge, no other study of MANETs has been performed so far with such a detailed physical layer model. Hence, this paper also gives new insights on the performance of MANETs in real outdoor environments.     

Optimization procedures for simultaneous road rehabilitation and bridge replacement decisions in highway networks

A road-bridge rehabilitation model is formulated as a mixed non-linear programming problem with linear constraints. The purpose of the model is to minimize user travel costs under limited availability of funds. Two constraints are related to budget availability for road rehabilitation and bridge replacement (or rehabilitation). For a given set of replaced bridges, the problem is reduced to a continuous non-linear programming problem that can be further decomposed into a traffic assignment problem (TAP) and a road rehabilitation budget allocation problem (RBAP). The solution to the non-linear problem is found by iteratively solving the TAP and the RBAP. Since the TAP has a non-convex objective function, its solution is only guaranteed to be a local optimum. Several local optima are obtained at each branch of the search tree to estimate a lower confidence limit on the user cost of a global solution.     

Text/shape classifier for mobile applications with handwriting input

The paper provides a practical solution to a real-time text/shape differentiation problem for online handwriting input. The proposed structure of the classification system comprises stroke grouping and stroke classification blocks. A new set of features is derived that has low computational complexity. The method achieves 98.5 % text/shape classification accuracy on a benchmark dataset. The proposed stroke grouping machine learning approach improves classification robustness in relation to different input styles. In contrast to the threshold-based techniques, this grouping adaptation enhances the overall discriminating accuracy of the text/shape recognition system by 11.3 %. The solution improves system’s response on a touch-screen device.     

Putting the 'P' into phospholipids

Despite almost hundred years of dedicated research, the synthesis of phospholipids remains a challenge. Here, we briefly review important trends and point out possible future directions that might lead to interesting new probes for the membrane environment. An improved liposome purification and concentration protocol rounds up the article.     

Multicriteria analysis based on constructing payoff matrices and applying methods of decision making in?fuzzy environment

There exist two classes of problems, which need the use of a multicriteria approach: problems whose solution consequences cannot be estimated with a single criterion and problems that, initially, may require a single criterion or several criteria, but their unique solutions are unachievable, due to decision uncertainty regions, which can be contracted using additional criteria. According to this, two classes of models (??X,M?? and ??X,R?? models) can be constructed. Analysis of ??X,M?? and ??X,R?? models (based on applying the Bellman-Zadeh approach to decision making in a fuzzy environment and using fuzzy preference modeling techniques, respectively) serves as parts of a general scheme for multicriteria decision making under information uncertainty. This scheme is also associated with a generalization of the classic approach to considering the uncertainty of information (based on analyzing payoff matrices constructed for different combinations of solution alternatives and states of nature) in monocriteria decision making to multicriteria problems. The paper results are of a universal character and are illustrated by an example.