Validation sets, genetic programming and generalisation

Jeannie Fitzgerald, Conor Ryan

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper investigates a new application of a validation set when using a three data set methodology with Genetic Programming (GP). Our system uses Validation Pressure combined with Validation Elitism to influence fitness evaluation and population structure with the aim of improving the system's ability to evolve individuals with an enhanced capacity for generalisation. This strategy facilitates the use of a validation set to reduce over-fitting while mitigating the loss of training data associated with traditional methods employing a validation set. The method is tested on five benchmark binary classification data sets and results obtained suggest that the strategy can deliver improved generalisation on unseen test data.

Original languageEnglish
Title of host publicationRes. and Dev. in Intelligent Syst. XXVIII
Subtitle of host publicationIncorporating Applications and Innovations in Intel. Sys. XIX - AI 2011, 31st SGAI Int. Conf. on Innovative Techniques and Applications of Artificial Intel.
PublisherSpringer London
Pages79-92
Number of pages14
ISBN (Print)9781447123170
DOIs
Publication statusPublished - 2011
Event1st SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence, AI 2011 - Cambridge, United Kingdom
Duration: 13 Dec 201115 Dec 2011

Publication series

NameRes. and Dev. in Intelligent Syst. XXVIII: Incorporating Applications and Innovations in Intel. Sys. XIX - AI 2011, 31st SGAI Int. Conf. on Innovative Techniques and Applications of Artificial Intel.

Conference

Conference1st SGAI International Conference on Innovative Techniques and Applications of Artificial Intelligence, AI 2011
Country/TerritoryUnited Kingdom
CityCambridge
Period13/12/1115/12/11

Fingerprint

Dive into the research topics of 'Validation sets, genetic programming and generalisation'. Together they form a unique fingerprint.

Cite this