Partial structure learning by subset Walsh transform.

Christie, Lee A.; Lonie, David P.; McCall, John A.W.

Abstract

Estimation of distribution algorithms (EDAs) use structure learning to build a statistical model of good solutions discovered so far, in an effort to discover better solutions. The non-zero coefficients of the Walsh transform produce a hypergraph representation of structure of a binary fitness function; however, computation of all Walsh coefficients requires exhaustive evaluation of the search space. In this paper, we propose a stochastic method of determining Walsh coefficients for hyperedges contained within the selected subset of the variables (complete local structure). This method also detects parts of hyperedges which cut the boundary of the selected variable set (partial structure), which may be used to incrementally build an approximation of the problem hypergraph.

Citation

CHRISTIE, L.A., LONIE, D.P. and MCCALL, J.A.W. 2013. Partial structure learning by subset Walsh transform. In Jin, Y. and Thomas, S.A. (eds.) Proceedings of the 13th UK workshop on computational intelligence (UKCI 2013), 9-11 September 2013, Guildford, UK. New York: IEEE [online], article number 6651297, pages 128-135. Available from: https://doi.org/10.1109/UKCI.2013.6651297

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 13th UK workshop on computational intelligence (UKCI 2013)
• Start Date: Sep 9, 2013
• End Date: Sep 11, 2013
• Acceptance Date: Sep 11, 2013
• Online Publication Date: Sep 11, 2013
• Publication Date: Oct 31, 2013
• Deposit Date: Feb 19, 2016
• Publicly Available Date: Feb 19, 2016
• Publisher: Institute of Electrical and Electronics Engineers (IEEE)
• Peer Reviewed: Peer Reviewed
• Article Number: 6651297
• Pages: 128-135
• Series Title: Proceedings of the UK workshop on computational intelligence
• ISBN: 9781479915682
• DOI: https://doi.org/10.1109/UKCI.2013.6651297
• Keywords: Walsh functions; Distributed algorithms; Graph theory; Set theory; Stochastic processes; Transforms; Computational modeling; Educational institutions; Equations; Estimation; Standards; Symmetric matrices; EDA; Walsh coefficients; Binary fitness function; Estimation of distribution algorithms; Hyperedges; Hypergraph representation; Nonzero coefficients; Partial structure learning; Statistical model; Stochastic method; Structure learning; Subset Walsh transform; Variable subset
• Public URL: http://hdl.handle.net/10059/1387
• Contract Date: Feb 19, 2016