Quantum Inspired Evolutionary Algorithm by Representing Candidate Solution as Normal Distribution
Lists
Only you see this
Abstract
Application of Quantum principles on evolutionary algorithms was started as early as late 1990s and has witnessed continued improvements since then. Following the same quantization principle introduced by the Quantum inspired evolutionary algorithm (QEA) in 2003, most of the existing quantum inspired algorithms focused mainly on evolving a single set of homogeneous solutions. In this paper, we present a new quantization process. In particular, aimed at solving numerical optimization problems, the evolutionary selection procedure is quantified through a set of subsolution points that jointly define candidate solutions. Implementing this new method on competitive co-evolution algorithm (CCEA), a new Quantum inspired competitive coevolution algorithm (QCCEA) is proposed in this paper. QCCEA is experimentally compared with CCEA through 9 benchmark numerical optimization functions published in CEC 2013. The results confirmed that QCCEA is more effective than CCEA over a majority of benchmark problems.
BibTeX
@inproceedings{tirumala2014quantum,
author = {Sreenivas Sremath Tirumala and Gang Chen and Shaoning Pang},
title = {{Quantum Inspired Evolutionary Algorithm by Representing Candidate Solution as Normal Distribution}},
booktitle = {ICONIP},
year = {2014},
url = {https://openreview.net/forum?id=Yb3xfqo6rac},
}Loading related content...