Convergence behavior of the (1 +, λ) evolution strategy on the ridge functions.

Oyman, Ahmet Irfan, Beyer, Hans-Georg, and Schwefel, Hans-Paul. "Convergence behavior of the (1 +, λ) evolution strategy on the ridge functions.." Mathware and Soft Computing 7.1 (2000): 35-75. <http://eudml.org/doc/39181>.

@article{Oyman2000,
abstract = {The convergence behavior of (1 +, λ)-ES is investigated at parabolic ridge, sharp ridge, and at the general case of the ridge functions. The progress rate, the distance to the ridge axis, the success rate, and the success probability are used in the analysis. The strong dependency of the (1 + λ)-ES to the initial conditions is shown using parabolic ridge test function when low distances to the ridge axis are chosen as the start value. The progress rate curve and the success probability curve of the sharp ridge is explained quite exactly using a simple local model. Two members of the corridor model family are compared to some members of the ridge function family, and they do not seem to be the limit case of the ridge function family according to our measures for convergence behavior.},
author = {Oyman, Ahmet Irfan, Beyer, Hans-Georg, Schwefel, Hans-Paul},
journal = {Mathware and Soft Computing},
keywords = {Modelo de evolución; Inteligencia artificial; Algoritmos genéticos; ridge functions},
language = {eng},
number = {1},
pages = {35-75},
title = {Convergence behavior of the (1 +, λ) evolution strategy on the ridge functions.},
url = {http://eudml.org/doc/39181},
volume = {7},
year = {2000},
}

TY - JOUR
AU - Oyman, Ahmet Irfan
AU - Beyer, Hans-Georg
AU - Schwefel, Hans-Paul
TI - Convergence behavior of the (1 +, λ) evolution strategy on the ridge functions.
JO - Mathware and Soft Computing
PY - 2000
VL - 7
IS - 1
SP - 35
EP - 75
AB - The convergence behavior of (1 +, λ)-ES is investigated at parabolic ridge, sharp ridge, and at the general case of the ridge functions. The progress rate, the distance to the ridge axis, the success rate, and the success probability are used in the analysis. The strong dependency of the (1 + λ)-ES to the initial conditions is shown using parabolic ridge test function when low distances to the ridge axis are chosen as the start value. The progress rate curve and the success probability curve of the sharp ridge is explained quite exactly using a simple local model. Two members of the corridor model family are compared to some members of the ridge function family, and they do not seem to be the limit case of the ridge function family according to our measures for convergence behavior.
LA - eng
KW - Modelo de evolución; Inteligencia artificial; Algoritmos genéticos; ridge functions
UR - http://eudml.org/doc/39181
ER -