Success History-Based Adaptive Differential Evolution Using Turning-Based Mutation

Single objective optimization algorithms are the foundation of establishing more complex methods, like constrained optimization, niching and multi-objective algorithms. Therefore, improvements to single objective optimization algorithms are important because they can impact other domains as well. This paper proposes a method using turning-based mutation that is aimed to solve the problem of premature convergence of algorithms based on SHADE (Success-History based Adaptive Differential Evolution) in high dimensional search space. The proposed method is tested on the Single Objective Bound Constrained Numerical Optimization (CEC2020) benchmark sets in 5, 10, 15, and 20 dimensions for all SHADE, L-SHADE, and jSO algorithms. The effectiveness of the method is verified by population diversity measure and population clustering analysis. In addition, the new versions (Tb-SHADE, TbL-SHADE and Tb-jSO) using the proposed turning-based mutation get apparently better optimization results than the original algorithms (SHADE, L-SHADE, and jSO) as well as the advanced DISH and the jDE100 algorithms in 10, 15, and 20 dimensional functions, but only have advantages compared with the advanced j2020 algorithm in 5 dimensional functions.

Download Full-text

Hybrid-adaptive differential evolution with decay function (HyDE-DF) applied to the 100-digit challenge competition on single objective numerical optimization

Proceedings of the Genetic and Evolutionary Computation Conference Companion on - GECCO '19 ◽

10.1145/3319619.3326747 ◽

2019 ◽

Cited By ~ 5

Author(s):

Fernando Lezama ◽

João Soares ◽

Ricardo Faia ◽

Zita Vale

Keyword(s):

Differential Evolution ◽

Numerical Optimization ◽

Decay Function ◽

Adaptive Differential Evolution ◽

Single Objective

Download Full-text

Insights into Exploration and Exploitation Power of Optimization Algorithm Using DSCTool

Mathematics ◽

10.3390/math8091474 ◽

2020 ◽

Vol 8 (9) ◽

pp. 1474

Author(s):

Peter Korošec ◽

Tome Eftimov

Keyword(s):

Statistical Analysis ◽

Web Service ◽

Optimization Algorithm ◽

Optimization Algorithms ◽

Search Space ◽

Exploration And Exploitation ◽

Learning Tool ◽

Additional Information ◽

E Learning ◽

Single Objective

When making statistical analysis of single-objective optimization algorithms’ performance, researchers usually estimate it according to the obtained optimization results in the form of minimal/maximal values. Though this is a good indicator about the performance of the algorithm, it does not provide any information about the reasons why it happens. One possibility to get additional information about the performance of the algorithms is to study their exploration and exploitation abilities. In this paper, we present an easy-to-use step by step pipeline that can be used for performing exploration and exploitation analysis of single-objective optimization algorithms. The pipeline is based on a web-service-based e-Learning tool called DSCTool, which can be used for making statistical analysis not only with regard to the obtained solution values but also with regard to the distribution of the solutions in the search space. Its usage does not require any special statistic knowledge from the user. The gained knowledge from such analysis can be used to better understand algorithm’s performance when compared to other algorithms or while performing hyperparameter tuning.

Download Full-text

An ancestor based extension to Differential Evolution (AncDE) for Single-Objective Computationally Expensive Numerical Optimization

2015 IEEE Congress on Evolutionary Computation (CEC) ◽

10.1109/cec.2015.7257293 ◽

2015 ◽

Cited By ~ 2

Author(s):

Rushikesh Sawant ◽

Donagh Hatton ◽

Diarmuid P. O'Donoghue

Keyword(s):

Differential Evolution ◽

Numerical Optimization ◽

Computationally Expensive ◽

Single Objective

Download Full-text

Design of a Parallel Manipulator Based on Multi-Objective Self-Adaptive Differential Evolution Algorithm

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.328.3 ◽

2013 ◽

Vol 328 ◽

pp. 3-8 ◽

Cited By ~ 1

Author(s):

Qing Mei Meng

Keyword(s):

Differential Evolution ◽

Differential Evolution Algorithm ◽

Search Space ◽

Parallel Robot ◽

Feasible Region ◽

Multi Objective ◽

Trial Vector ◽

Adaptive Differential Evolution ◽

Optimal Pid Control ◽

Self Adaptive

In order to improve highly non-isotropic input-output relations in the optimal design of a parallel robot, this paper presents a method based on a multi-objective self-adaptive differential evolution (MOSaDE) algorithm.The approach considers a solution-diversity mechanism coupled with a memory of those sub-optimal solutions found during the process. In theMOSaDE algorithm, both trial vector generation strategies and their associated control parameter values were gradually self-adapted by learning from their previous experiences in generating promising solutions. Consequently, a more suitable generation strategy along with its parameter settings could be determined adaptively to match different phases of the search processevolution.Furthermore, a constraint-handling mechanism is added to bias the search to the feasible region of the search space. The obtained solution will be a set of optimal geometric parameters and optimal PID control gains. The empirical analysis of thenumerical results shows the efficiency of the proposed algorithm.

Download Full-text

The Hybrid Differential Evolution with Dynamic Scaling Mutation and Wrapper Local Search for Optimization Problems

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.479-480.989 ◽

2013 ◽

Vol 479-480 ◽

pp. 989-995

Author(s):

Chun Liang Lu ◽

Shih Yuan Chiu ◽

Chih Hsu Hsu ◽

Shi Jim Yen

Keyword(s):

Local Search ◽

Differential Evolution ◽

Job Shop ◽

Job Shop Scheduling ◽

Optimization Problems ◽

Search Space ◽

Dynamic Scaling ◽

Multi Objective ◽

Wide Range ◽

Single Objective

In this paper, an improved hybrid Differential Evolution (DE) is proposed to enhance optimization performance by cooperating Dynamic Scaling Mutation (DSM) and Wrapper Local Search (WLS) schemes. When evolution speed is standstill, DSM can improve searching ability to achieve better balance between exploitation and exploration in the search space. Furthermore, WLS can disturb individuals to fine tune the searching range around and then properly find better solutions in the evolution progress. The effective particle encoding representation named Particle Segment Operation-Machine Assignment (PSOMA) that we previously published is also applied to always produce feasible candidate solutions for hybrid DE model to solve the Flexible Job-Shop Scheduling Problem (FJSP). To test the performance of the proposed hybrid method, the experiments contain five frequently used CEC 2005 numerical functions and three representative FJSP benchmarks for single-objective and multi-objective optimization verifications, respectively. Compare the proposed method with the other related published algorithms, the simulation results indicate that our proposed method exhibits better performance for solving most the test functions for single-objective problems. In addition, the wide range of Pareto-optimal solutions and the more Gantt chart diversities can be obtained for the multi-objective FJSP in practical decision-making considerations.

Download Full-text