TY - GEN
T1 - The Initialization of Evolutionary Multi-objective Optimization Algorithms
AU - Hamdan, Mohammad
AU - Qudah, Osamah
PY - 2015
Y1 - 2015
N2 - Evolutionary algorithms are the most widely used meta-heuristics for solving multi objective optimization problems, and since all of these algorithms are population based, such as NSGAII, there are a set of factors that affect the final outcomes of these algorithms such as selection criteria, crossover, mutation and fitness evaluation. Unfortunately, little research sheds light at how to generate the initial population. The common method is to generate the initial population randomly. In this work, a set of initialization methods were examined such as, Latin hypercube sampling (LHS), Quasi-Random sampling and stratified sampling. Nonetheless. We also propose a modified version of Latin Hypercube sampling method called (Quasi_LHS) that uses Quasi random numbers as a backbone in its body. Furthermore, we propose a modified version of Stratified sampling method that uses Quasi-Random numbers to represent the intervals. For our research, a set of well known multi objective optimization problems were used in order to evaluate our initial population strategies using NSGAII algorithm. The results show that the proposed initialization methods (Quasi_LHS) and Quasi-based Stratified improved to some extent the quality of final results of the experiments.
AB - Evolutionary algorithms are the most widely used meta-heuristics for solving multi objective optimization problems, and since all of these algorithms are population based, such as NSGAII, there are a set of factors that affect the final outcomes of these algorithms such as selection criteria, crossover, mutation and fitness evaluation. Unfortunately, little research sheds light at how to generate the initial population. The common method is to generate the initial population randomly. In this work, a set of initialization methods were examined such as, Latin hypercube sampling (LHS), Quasi-Random sampling and stratified sampling. Nonetheless. We also propose a modified version of Latin Hypercube sampling method called (Quasi_LHS) that uses Quasi random numbers as a backbone in its body. Furthermore, we propose a modified version of Stratified sampling method that uses Quasi-Random numbers to represent the intervals. For our research, a set of well known multi objective optimization problems were used in order to evaluate our initial population strategies using NSGAII algorithm. The results show that the proposed initialization methods (Quasi_LHS) and Quasi-based Stratified improved to some extent the quality of final results of the experiments.
KW - Evolutionary algorithms
KW - Initial population
KW - Latin hypercube sampling
KW - NSGAII algorithm
KW - Pareto set
KW - Quality indicators
KW - Quasi random numbers
KW - Random numbers
KW - Stratified sampling
UR - http://www.scopus.com/inward/record.url?scp=84947813837&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-20466-6_52
DO - 10.1007/978-3-319-20466-6_52
M3 - Conference contribution
AN - SCOPUS:84947813837
SN - 9783319204659
T3 - Lecture Notes in Computer Science
SP - 495
EP - 504
BT - Advances in Swarm and Computational Intelligence
A2 - Tan, Ying
A2 - Shi, Yuhui
A2 - Buarque, Fernando
A2 - Gelbukh, Alexander
A2 - Das, Swagatam
A2 - Engelbrecht, Andries
PB - Springer
T2 - 6th International Conference on Swarm Intelligence 2015
Y2 - 25 June 2015 through 28 June 2015
ER -