We consider evolutionary model selection for support vector machines. Hold-out set-based objective functions are natural model selection criteria, and we introduce a symmetrization of the standard cross-validation approach. We propose the covariance matrix adaptation evolution strategy (CMA-ES) with uncertainty handling for optimizing the new randomized objective function. Our results show that...

Inspired by the phenomena of evolution in nature and society, this paper put forward a new model for evolutionary algorithms to conquer its limitation of premature convergence. Then the model is used to optimize the parameters of model of crop in greenhouse and compared the result to that of the simple Evolutionary Algorithms under the same situation. Finally it is convinced that the new model ...

This paper presents an evolutionary algorithm for solving parameter optimization problems in parametric array signal processing. The method utilises the niche concept to maintain a population capable of exploring the search space while preserving the diversity to prevent the premature convergence. The former is achieved by a niche-based adaptive mutation scheme and the latter by a niche-based l...

An algorithm for identifying parameters in dynamical systems is developed in this work using homotopy transformations and the single-shooting method. The equations governing the dynamics of the mathematical model are augmented with observer-like homotopy terms that smooth the objective function. As a result, premature convergence to a local minimum is avoided and the obtained parameter estimate...

We explore the advantage of using genetic algorithms for the automated design of a small satellite passive thermal control. The control law is a layout of heterogeneous material patches covering the satellite surface. We tested our candidate material patch solutions with a finite element method simulation of the satellite thermal conditions during direct sun light orbit. We found that the metho...

For most practical purposes, true topology optimization of a braced frame should be synchronized with its sizing. An integrated layout optimization is formulated here to simultaneously account for both member sizing and bracings’ topology in such a problem. Code-specific seismic design spectrum is applied to unify the earthquake excitation. The problem is solved for minimal structural weight un...

Bilevel programming problem (BLP) is a nested optimization problem that contains one optimization task as a constraint to another optimization task. In this paper, a hybrid differential evolution and particle swarm (DE-PSO) is proposed for solving the linear BLP, in which the hybrid strategy can efficiently prevent the premature convergence of the swarm. Finally, we use the test problems from t...

The popularity of Ontology favored the appearance of several Ontologies to the same domain, thereby increasing the need of alignment techniques. In scenarios where ontologies comprising instances, the knowledge embedded in these instances can be useful to improve alignment. This paper extends a hybrid evolutionary approach, which applies a local optimization method, by taking instances into acc...

In continuous black-box optimization, various stochastic local search techniques are often employed, with various remedies for fighting the premature convergence. This paper surveys recent developments in the field (the most important from the author’s perspective), analyzes the differences and similarities and proposes a taxonomy of these methods. Based on this taxonomy, a variety of novel, pr...

Immune algorithms (IAs) are microscopic view of evolutionary algorithms (EAs) and applied in combinatorial optimization problems. This paper addresses to a clonal selection algorithm (CSA) that is one of the most representative IA and was applied into the protein structure prediction (PSP) on AB off-lattice model, in which the memory B cells of the CSA was innovated by employing different strat...