Toggle Main Menu Toggle Search

Open Access padlockePrints

Automatic Tuning of Rule-Based Evolutionary Machine Learning via Problem Structure Identification

Lookup NU author(s): Dr Maria Franco Gaviria, Professor Natalio KrasnogorORCiD, Professor Jaume Bacardit

Downloads


Licence

This is the authors' accepted manuscript of an article that has been published in its final definitive form by IEEE, 2020.

For re-use rights please refer to the publisher's terms and conditions.


Abstract

© 2005-2012 IEEE. The success of any machine learning technique depends on the correct setting of its parameters and, when it comes to large-scale datasets, hand-tuning these parameters becomes impractical. However, very large-datasets can be pre-processed in order to distil information that could help in appropriately setting various systems parameters. In turn, this makes sophisticated machine learning methods easier to use to end-users. Thus, by modelling the performance of machine learning algorithms as a function of the structure inherent in very large datasets one could, in principle, detect "hotspots"in the parameters' space and thus, auto-tune machine learning algorithms for better dataset-specific performance. In this work we present a parameter setting mechanism for a rule-based evolutionary machine learning system that is capable of finding the adequate parameter value for a wide variety of synthetic classification problems with binary attributes and with/without added noise. Moreover, in the final validation stage our automated mechanism is able to reduce the computational time of preliminary experiments up to 71% for a challenging real-world bioinformatics dataset.


Publication metadata

Author(s): Franco MA, Krasnogor N, Bacardit J

Publication type: Article

Publication status: Published

Journal: IEEE Computational Intelligence Magazine

Year: 2020

Volume: 15

Issue: 3

Pages: 28-46

Print publication date: 01/08/2020

Online publication date: 15/07/2020

Acceptance date: 02/04/2018

Date deposited: 06/11/2020

ISSN (print): 1556-603X

ISSN (electronic): 1556-6048

Publisher: IEEE

URL: https://doi.org/10.1109/MCI.2020.2998232

DOI: 10.1109/MCI.2020.2998232


Altmetrics

Altmetrics provided by Altmetric


Funding

Funder referenceFunder name
EP/M020576/1EPSRC
EP/N031962/1EPSRC
EP/H016597/1
EPSRC

Share