Narrow Results

This paper presents a structural distance-based crossover for neural network classifiers, which is applied as part of a Memetic Algorithm (MA) for evolving simultaneously the structure and weights of neural network models applied to multiclass problems. Previous researchers have shown that this simultaneous evolution is a way to avoid the noisy fitness evaluation. The MA incorporates a crossover operator that shows to be useful for ameliorating the permutation problem of the network representation (i.e. different genotypes can be used to represent the same neural network phenotype), increasing the structural diversity of the individuals and improving the accuracy of the results. Instead of a recombination probability, the crossover operator considers a similarity parameter (the minimum structural distance), which allows to maintain a trade-off between global and local search. The neural network models selected in this work are the product-unit neural networks (PUNNs), due to their increasing relevance in those classification problems which show a high order relationship between the input variables. The proposed MA is intended to reduce the possible overtraining problems which can raise in some datasets for this kind of models. The evolutionary system is applied to eight classification benchmarks and the results of an analysis of variance contrast (ANOVA) show the effectiveness of the structural-based crossover operator and the capacity of our algorithm to obtain evolved PUNNs with a higher classification accuracy than those obtained using other evolutionary techniques. On the other hand, the results obtained are compared with popular effective machine learning classification methods, resulting in a competitive performance.

Optimal sensor placement (OSP) method plays a key role in setting up a health monitoring system for large-scale structures. This paper describes the implementation of monkey algorithm (MA) as a strategy for the optimal placement of a predefined number of sensors. To effectively maintain the population diversity while enhancing the exploitation capacities during the optimization process, a novel niching monkey algorithm (NMA) by combining the MA with the niching techniques is developed in this paper. In the NMA, the dual-structure coding method is adopted to code the design variables and a chaos-based approach instead of a pure random initialization is employed to initialize the monkey population. Meanwhile, the niche generation operation and fitness sharing mechanism are modified and incorporated to alleviate the premature convergence problem while enhancing the exploration of new search domain. In addition, to promote interactions and share the available resources, the replacement scheme is proposed and adopted among the niches. Finally, numerical experiments are conducted on a high-rise structure to evaluate the performance of the proposed NMA. It is found that the innovations in the proposed NMA can effectively improve the convergence of algorithm and generate superior sensor configurations when compared to the original MA.