Please use this identifier to cite or link to this item:
Appears in Collections:Computing Science and Mathematics Journal Articles
Peer Review Status: Refereed
Title: Learning and Searching Pseudo-Boolean Surrogate Functions from Small Samples
Author(s): Swingler, Kevin
Keywords: Fitness Function Modelling
Estimation of Distribution Algorithms
Pseudo-Boolean Functions
Linkage Learning
Walsh Decomposition
Mixed Order Hyper Networks
Statistical Machine Learning
Issue Date: 2020
Date Deposited: 30-May-2019
Citation: Swingler K (2020) Learning and Searching Pseudo-Boolean Surrogate Functions from Small Samples. Evolutionary Computation, 28 (2), pp. 317-338.
Abstract: When searching for input configurations that optimise the output of a system, it can be useful to build a statistical model of the system being optimised. This is done in approaches such as surrogate model-based optimisation, estimation of distribution algorithms and linkage learning algorithms. This paper presents a method for modelling pseudo-Boolean fitness functions using Walsh bases and an algorithm designed to discover the non-zero coefficients while attempting to minimise the number of fitness function evaluations required. The resulting models reveal linkage structure that can be used to guide a search of the model efficiently. It presents experimental results solving benchmark problems in fewer fitness function evaluations than those reported in the literature for other search methods such as EDAs and linkage learners.
DOI Link: 10.1162/evco_a_00257
Rights: Accepted for publication in Evolutionary Computation published by MIT Press. The final published version is available at:

Files in This Item:
File Description SizeFormat 
ECJ-2018-036R2-single.pdfFulltext - Accepted Version207.02 kBAdobe PDFView/Open

This item is protected by original copyright

Items in the Repository are protected by copyright, with all rights reserved, unless otherwise indicated.

The metadata of the records in the Repository are available under the CC0 public domain dedication: No Rights Reserved

If you believe that any material held in STORRE infringes copyright, please contact providing details and we will remove the Work from public display in STORRE and investigate your claim.