2026-04-14T20:59:25Z https://meral.edu.mm/oai

oai:meral.edu.mm:recid/4767 2021-12-13T06:20:36Z 1582963302567:1597824273898 user-ucsy

Parallel Differential Evolution Algorithm with Multiple Trial Vectors to Artificial Neural Network Training Thein, Htet Thazin Tike Tun, Khin Mo Mo In this paper, parallel differential evolutionalgorithm with multiple trial vectors for trainingartificial neural networks (ANNs) is presented. Theproposed method is PDEA, which is a DE-ANN+modified by adding island model. Within PDEA, anisland model is designed to cooperatively search forthe global optima in search space. By combining thestrengths of the differential evolution algorithm withmultiple trial vectors and island model, PDEA greatlyimproves the optimization performance. PDEAalgorithm is used for ANN training to classify theparity-p problem. Results obtained using proposedalgorithm has been compared to the results obtainedusing other evolutionary algorithms. 2015-02-05 http://hdl.handle.net/20.500.12678/0000004767 https://meral.edu.mm/records/4767