Search

The expensive optimization problem, widespread in real-world applications, poses great challenges to traditional evolutionary algorithms (EAs) due to its costly evaluations. A lot of investigations have proved that it is a practical way to incorporate surrogate models into EAs to reduce the number of time-consuming evaluations. Although various types of surrogate models from the community of machine learning, such as regression models that approximate the fitness values of candidate solutions or classification models that predict candidate solutions' category, have been widely applied separately, none of them consistently outperforms the other. Considering the intrinsic consistency of alleviating computational burden to improve the optimization efficiency, this paper introduces a two-layer architecture to integrate regression and classification models and synergize global and local evolution for high-dimensional expensive many-objective optimization problem (EMaOP). To implement this idea, the Kriging model based on incremental learning is utilized to explore the potential promising space. Further, multi local support vector machine (SVM) classifiers assisted improved differential evolution is run by turn with the global search to accelerate convergence. Accordingly, a novel surrogate-assisted EA, named Classification- and Regression-assisted evolutionary algorithm (CR-SAEA), is proposed on this basis. Experimental studies on several benchmark functions with up to 11 objectives and 200 dimensions confirm the effectiveness of the proposed algorithm and its competitiveness against state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

The paper aims to solve the problem of multi-agent path planning in complex environment using optimization algorithm. To address the issue of local optimum and premature convergence, a new method is proposed based on the whale optimization algorithm, combining the chaotic initialization, the reverse search and the differential evolution methods. It is theoretically proved that this algorithm is globally convergent in probability. When applied to path planning problems, the proposed optimization algorithm can effectively find a globally optimal and smoother path. Through simulation experiments with multi-UAVs, it is demonstrated that the proposed algorithm has better performance than the state-of-the-art methods in environment with both static and dynamic obstacles, reflecting the global convergence and robustness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

An image is often corrupted by different kinds of noise during its acquisition and transmission. Conventional denoising methods can suppress the Gaussian noise effectively, but fail to maintain the quality of denoised images and may blur edges in an image. To address these short comings, this paper aims to develop an optimized adaptive thresholding function based framework for edge preserved satellite image denoising using different nature inspired algorithms which is capable of effectively removing the Gaussian noise from images without over smoothing edge details. Image denoising using adaptive thresholding functions selects the suitable threshold values to separate noise from the actual image without affecting the actual features of the image. In this approach, most widely used nature inspired optimization algorithms are exploited for learning the parameters of adaptive thresholding function required for optimum performance. It was found that the proposed adaptive differential evolution algorithm (JADE) algorithm based denoising approach has superior features and give better performance in terms of PSNR, MSE, SSIM and FSIM as compared to other methods. [ABSTRACT FROM AUTHOR]