Application Of Evolutionary Multi-objective Optimization Research Articles

Evolutionary multi-objective automatic clustering enhanced with quality metrics and ensemble strategy

Automatic clustering problem, which needs to detect the appropriate clustering without a pre-defined number of clusters (k), is difficult and challenging in unsupervised learning owing to the lack of prior domain knowledge. Despite a rising tendency with the application of evolutionary multi-objective optimization (EMO) techniques for automatic clustering, there still exist some obvious under-explored issues. In this paper, we resort to quality metrics and ensemble strategy for the sake of explicit/implicit knowledge discovery to guide the optimization process. The quality and diversity of solutions defined in terms of cluster validities, as similar to performance indicator for multi-objective optimization, are applied to assist in addressing automatic clustering problems and decreasing unnecessary computational overhead. To be specific, the main components like initialization, reproduction operations, and environmental selection which involved during EMO based automatic clustering are discussed and refined. For the determination of the final partitioning, quality metrics and cluster ensemble strategy are both considered to improve the retrieve system in the unsupervised way. Experiments are conducted from several different aspects and the corresponding analyses are provided, which confirm that the proposals are more efficient and effective for automatic clustering.

Using low-power platforms for Evolutionary Multi-Objective Optimization algorithms

Nowadays, the application of Evolutionary Multi-Objective Optimization (EMO) algorithms in real-time systems receives considerable interest. In this context, the energy efficiency of computational systems is of paramount relevance. Recently, the use of embedded systems based on heterogeneous (CPU + GPU) platforms is consistently increasing. For example, NVIDIA Jetson cards are low-power computers designed for development of embedded applications. They incorporate Tegra processors which feature a CUDA-capable GPU. This way, Jetson cards can be considered as a prototype of low-power computer of High-Performance Computing. In this work, our interest is focused on the NSGA-II algorithm, a well-known representative of EMO algorithms. The strength of NSGA-II lies in its Non-Dominated Sorting (NDS) procedure of a population of individuals. Our purpose on the low-power computers is twofold: to define and evaluate the parallel NSGA-II versions with major focus on NDS procedure on the Jetson platforms and to determinate the size of NSGA-II problems which can be solved. The results show that the parallel version which achieves the best performance depends on the objectives functions and the frequencies of the clocks of the cores and memory of the GPU. The analysis of the results shows the capability of the Jetson as a low-consumption platform which allows to accelerate the execution of instances of the state-of-the-art EMO algorithm--NSGA-II.

Predicting DNA Motifs by Using Evolutionary Multiobjective Optimization

Bioinformatics and computational biology include researchers from many areas: biochemists, physicists, mathematicians, and engineers. The scale of the problems that are discussed ranges from small molecules to complex systems, where many organisms coexist. However, among all these issues, we can highlight genomics, which studies the genomes of microorganisms, plants, and animals. Predicting common patterns, i.e., motifs, in a set of deoxyribonucleic acid (DNA) sequences is one of the important sequence analysis problems, and it has not yet been resolved in an efficient manner. In this study, we study the application of evolutionary multiobjective optimization to solve the motif discovery problem, applied to the specific task of discovering novel transcription factor binding sites in DNA sequences. For this, we have designed, adapted, configured, and evaluated several types of multiobjective metaheuristics. After a detailed study, the results indicate that these metaheuristics are appropriate for discovering motifs. To find good approximations to the Pareto front, we use the hypervolume indicator, which has been successfully integrated into evolutionary algorithms. Besides the hypervolume indicator, we also use the coverage relation to ensure: Which is the best Pareto front? New results have been obtained, which significantly improve those published in previous research works.

Multidisciplinary Design Optimization and Data Mining for Transonic Regional-Jet Wing

A large-scale, real-world application of evolutionary multi-objective optimization is reported. The multidisciplinary design optimization among aerodynamics, structures, and aeroelasticity of the wing of a transonic regional-jet aircraft was performed using high-fidelity evaluation models. Euler and Navier-Stokes solvers were employed for aerodynamic evaluation. The commercial software NASTRAN was coupled with a computational fluid dynamics solver for the structural and aeroelastic evaluations. An adaptive range multi-objective genetic algorithm was employed as an optimizer. The objective functions were minimizations of block fuel and maximum takeoff weight in addition to drag divergence between transonic and subsonic flight conditions. As a result, nine nondominated solutions were generated and used for tradeoff analysis among three objectives. Moreover, all solutions evaluated during the evolution were analyzed using a self-organizing map as a data mining technique to extract key features of the design space. One of the key features found by data mining was the nongull wing geometry, although the present multidisciplinary design optimization results showed the inverted gull wings as nondominated solutions. When this knowledge was applied to one optimum solution, the resulting design was found to have better performance and to achieve 3.6% improvement in the block fuel compared to the original geometry designed in the conventional manner.

Evolutionary multiobjective optimization for the design of fuzzy rule-based ensemble classifiers

In this paper, we examine the application of evolutionary multiobjective optimization (EMO) algorithms to the design of fuzzy rule-based ensemble classifiers. An EMO algorithm is used to search for a large number of non-dominated fuzzy rule-based cla

Procedural texture evolution using multi-objective optimization

This paper investigates the application of evolutionary multiobjective optimization to two-dimensional procedural texture synthesis. Genetic programming is used to evolve procedural texture formulae. Earlier work used multiple feature tests during fitness evaluation to rate how closely a candidate texture matches visual characteristics of a target texture image. These feature test scores were combined into all overall fitness score using a weighted sum. This paper improves this research by replacing the weighted sum with a Pareto ranking scheme, which preserves the independence of feature tests during fitness evaluation. Three experiments were performed: a pure Pareto ranking scheme, and two Pareto experiments enhanced with parameterless population divergence strategies. One divergence strategy is similar to that used by the NSGA-II system, and scores individuals using their nearest-neighbour distance in feature-space. The other strategy uses a normalized, ranked abstraction of nearest neighbour distance. A result of this work is that acceptable textures can be evolved much more efficiently and with less user intervention with MOP evolution than compared to the weighted sum approach. Although the final acceptability of a texture is ultimately a subjective decision of the user, the proposed use of multi-objective evolution is useful for generating for the user a diverse assortment of possibilities that reflect the important features of interest.