An Application of Tree Seed Algorithm on Optimizing 50 and 100 Dimensional Numeric Functions

EasyChair Preprint 2685

Abstract

The Tree-Seed Algorithm, TSA for short, is a meta-heuristic optimization algorithm inspired by the relationships between trees and seeds. The performance of TSA on the lower dimensional function optimization had been proved, in this work, we applied TSA to optimize higher dimensional functions. In this study, a version of TSA has been developed, called iTSA, which has been applied to solve 50 and 100 dimensional numeric functions. The improvement is based on the usage of more solution update mechanisms instead of one mechanism. In experiments, CEC2015 benchmark functions are used and iTSA is compared with the basic version of TSA, artificial bee colony, particle swarm optimization and some variants of differential evolution algorithm. The experimental results are reported as mean, max, min solutions and standard deviation of the 30 different runs. The experimental results also show that the proposed algorithm produces comparable and robust solution in terms of solution quality and robustness.

Keyphrases: CEC2015, Tree-Seed Algorithm, Update mechanism, function optimization

@booklet{EasyChair:2685,
  author    = {Mustafa Servet Kiran and Imral Güngör},
  title     = {An Application of Tree Seed Algorithm on Optimizing 50 and 100 Dimensional Numeric Functions},
  howpublished = {EasyChair Preprint 2685},
  year      = {EasyChair, 2020}}