Hierarchical surrogate-assisted evolutionary optimization framework

Zhou, Zongzhao, Ong, Yew Soon and Nair, P.B. (2004) Hierarchical surrogate-assisted evolutionary optimization framework. In, Congress on evolutionary computing. Congress on Evolutionary Computing 2004 , Institute of Electrical and Electronics Engineers, 1586-1593. (doi:10.1109/CEC.2004.1331085).


PDF - Accepted Manuscript
Download (1732Kb)
Original Publication URL: http://dx.doi.org/10.1109/CEC.2004.1331085


This work presents enhancements to a surrogate-assisted evolutionary optimization framework proposed earlier in the literature for solving computationally expensive design problems on a limited computational budget (Ong et al., 2003). The main idea of our former framework was to couple evolutionary algorithms with a feasible sequential quadratic programming solver in the spirit of Lamarckian learning, including a trust-region approach for interleaving the true fitness function with computationally cheap local surrogate models during gradient-based search. We propose a hierarchical surrogate-assisted evolutionary optimization framework for accelerating the convergence rate of the original surrogate-assisted evolutionary optimization framework. Instead of using the exact high-fidelity fitness function during evolutionary search, a Kriging global surrogate model is used to screen the population for individuals that undergo Lamarckian learning. Numerical results are presented for two multimodal benchmark test functions to show that the proposed approach leads to a further acceleration of the evolutionary search process.

Item Type: Book Section
Digital Object Identifier (DOI): doi:10.1109/CEC.2004.1331085
Additional Information: CEC2004
Related URLs:
Keywords: evolutionary computation, gradient methods, learning (artificial intelligence), quadratic programming, search problems, statistical analysis
Subjects: T Technology
Q Science > QA Mathematics > QA76 Computer software
Divisions : University Structure - Pre August 2011 > School of Engineering Sciences
ePrint ID: 22967
Accepted Date and Publication Date:
Date Deposited: 03 Apr 2006
Last Modified: 31 Mar 2016 11:43
URI: http://eprints.soton.ac.uk/id/eprint/22967

Actions (login required)

View Item View Item

Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.

View more statistics