Research Article
MVMO with Opposite Gradient Initialization for Single Objective Problems
@INPROCEEDINGS{10.1007/978-3-030-06152-4_11, author={Thirachit Saenphon}, title={MVMO with Opposite Gradient Initialization for Single Objective Problems}, proceedings={Context-Aware Systems and Applications, and Nature of Computation and Communication. 7th EAI International Conference, ICCASA 2018, and 4th EAI International Conference, ICTCC 2018, Viet Tri City, Vietnam, November 22--23, 2018, Proceedings}, proceedings_a={ICCASA \& ICTCC}, year={2019}, month={1}, keywords={Continuous function Mean-variance mapping optimization Opposite gradient initialization search Optimization}, doi={10.1007/978-3-030-06152-4_11} }- Thirachit Saenphon
Year: 2019
MVMO with Opposite Gradient Initialization for Single Objective Problems
ICCASA & ICTCC
Springer
DOI: 10.1007/978-3-030-06152-4_11
Abstract
The objective of this paper is to describe an opposite gradient initialization concept with mean-variance mapping optimization (OGI-MVMO). OGI-MVMO is an optimization based on the actual manifold of objective function whereas original MVMO based stochastic optimization. Generating the new candidate solution to speed up the solution finding and accuracy of solution are important purposes. The OGI-MVMO algorithm consist of 2 steps: the primary step is generating new solution by OGI and also the second step is mutation between every of selected candidate solution supported the mean and variance of the population. The results showed that OGI-MVMO algorithm has better performance than other algorithm include the original MVMO for 15 real-parameter single objective functions.
