Discovering predictive variables when evolving cognitive models

P.C.R. Lane, F. Gobet

Research output: Contribution to journalArticlepeer-review

10 Citations (Scopus)
34 Downloads (Pure)

Abstract

A non-dominated sorting genetic algorithm is used to evolve models of learning from di erent theories for multiple tasks. Correlation analysis is performed to identify parameters which affect performance on specific tasks; these are the predictive variables. Mutation is biased so that changes to parameter values tend to preserve values within the population's current range. Experimental results show that optimal models are evolved, and also that uncovering predictive variables is beneficial in improving the rate of convergence.
Original languageEnglish
Pages (from-to)108-117
JournalLecture Notes in Computer Science (LNCS)
Issue number3rd Int Conf on Advances in Pattern Recognition
DOIs
Publication statusPublished - 2005

Fingerprint

Dive into the research topics of 'Discovering predictive variables when evolving cognitive models'. Together they form a unique fingerprint.

Cite this