Analysis of a master-slave architecture for distributed evolutionary computations.

This paper introduces a new mathematical model of the master-slave architecture for distributed evolutionary computations (EC). This model is validated using a concrete implementation based on the Distributed BEAGLE C++ framework. Results show that contrary to (current) popular belief, master-slave architectures are able to scale well over local area networks of workstations using off-the-shelf networking equipment. The main properties of the master-slave are also compared with those of the more mainstream island-model.     

Training hidden Markov models with multiple observations-acombinatorial method

Hidden Markov models (HMM) are stochastic models capable of statistical learning and classification. They have been applied in speech recognition and handwriting recognition because of their great adaptability and versatility in handling sequential signals. On the other hand, as these models have a complex structure and also because the involved data sets usually contain uncertainty, it is difficult to analyze the multiple observation training problem without certain assumptions. For many years researchers have used the training equations of Levinson (1983) in speech and handwriting applications, simply assuming that all observations are independent of each other. This paper presents a formal treatment of HMM multiple observation training without imposing the above assumption. In this treatment, the multiple observation probability is expressed as a combination of individual observation probabilities without losing generality. This combinatorial method gives one more freedom in making different dependence-independence assumptions. By generalizing Baum's auxiliary function into this framework and building up an associated objective function using the Lagrange multiplier method, it is proven that the derived training equations guarantee the maximization of the objective function. Furthermore, we show that Levinson's training equations can be easily derived as a special case in this treatment     

Genetic engineering of hierarchical fuzzy regional representations for handwritten character recognition

This paper presents a genetic programming based approach for optimizing the feature extraction step of a handwritten character recognizer. This recognizer uses a simple multilayer perceptron as a classifier and operates on a hierarchical feature space of orientation, curvature, and center of mass primitives. The nodes of the hierarchy represent rectangular sub-regions of their parent node, the tree root corresponding to the character's bounding box. Within each sub-region, a variable number of fuzzy features are extracted. Genetic programming is used to simultaneously learn the best hierarchy and the best combination of fuzzy features. Moreover, the fuzzy features are not predetermined, they are inferred from the evolution process which runs a two-objective selection operator. The first objective maximizes the recognition rate, and the second minimizes the feature space size. Results on Unipen data show that, using this approach, robust representations could be obtained that out-performed comparable human designed hierarchical fuzzy regional representations.     

Human-competitive lens system design with evolution strategies

Lens system design provides ideal problems for evolutionary algorithms: a complex non-linear optimization task, often with intricate physical constraints, for which there is no analytical solutions. This paper demonstrates, through the use of two evolution strategies, namely non-isotropic Self-Adaptive evolution strategy (SA-ES) and Covariance Matrix Adaptation evolution strategy (CMA-ES), as well as multiobjective Non-Dominated Sort Genetic Algorithm 2 (NSGA-II) optimization, the human competitiveness of an approach where an evolutionary algorithm is hybridized with a local search algorithm to solve both a classic benchmark problem, and a real-world problem.     

Using Evolutionary Optimization Techniques for Scheduling Water Pipe Renewal Considering a Short Planning Horizon

Abstract: The maintenance and management of underground infrastructures is a growing problem for a majority of municipalities. The maintenance costs are increasing while the financial resources of municipalities remain limited. Water distribution system (WDS) managers therefore need tools to assist them in the elaboration of pipe renewal management plans. In this article, results of a newly developed strategy for pipe renewal based on a cost function are presented. The strategy allows the minimization of a cost function while also considering hydraulic criterion. This strategy was tested on a short planning horizon of five years. The pipe number to be replaced and the optimal moment for renewal are identified using three different optimization techniques: IGA (Island Genetic Algorithm), NPGA‐2 (Niched Pareto Genetic Algorithm 2), and NSGA‐II (Non‐dominated Sorting Genetic Algorithm‐II). The proposed approach has five distinctive features: (1) it is coupled with a flexible evolutionary framework that allows the user to select any type of operator for IGA or any kind of multiobjective genetic algorithm; (2) it uses the hydraulic simulator Epanet2.0 which allows steady state or dynamic simulations; (3) it considers a probabilistic break model to evaluate the structural deterioration of pipes; (4) it integrates a Bayesian approach for the estimation of the pipe break model parameters that take into account the influence of inherent uncertainties related to the quality of data during the decision‐making process; and (5) it simulates the variation of the pipe's roughness over the years. The developed strategy/model is explained using an example that allows us to elucidate its most important components. Simulation experiments on a small network (100 pipes) are presented. A comparison of three evolutionary algorithm results is provided. Tests showed that IGA performs well, but for networks of larger sizes, we recommend increasing the number of demes to reach better solutions. Higher quality results were achieved with NSGA‐II than NPGA‐2 on differently sized networks. We recommend the use the NSGA‐II to optimize large WDS. Future developments for this strategy are also discussed.