Vis enkel innførsel

dc.contributor.authorHoueland, Tor Gunnar Høst
dc.date.accessioned2020-06-09T06:09:25Z
dc.date.available2020-06-09T06:09:25Z
dc.date.issued2020
dc.identifier.isbn978-82-326-4735-4
dc.identifier.issn1503-8181
dc.identifier.urihttps://hdl.handle.net/11250/2657253
dc.description.abstractMachine learning systems are becoming increasingly widespread and important, and these days machine learning is used in some form in most industries. However, the application of machine learning technology still has a relatively high barrier to entry, requiring both machine learning expertise and domain knowledge. In this thesis, we present a metareasoning approach to multi-method machine learning that allows the system to adapt and optimize learning for a given domain automatically, without requiring human expert judgment. In contrast to popular deep learning methods for similar situations, the approach presented here does not require specialized hardware nor large data sets. Multiple machine learning components are continuously evaluated at run-time while solving problems, using a framework to analyze overall system performance based on observed prediction performance and time spent. The system automatically learns to prioritize the methods with the best empirical performance for a given domain. In experiments using data sets from the UCI machine learning repository and machine learning methods from the Weka suite, an example implementation outperformed individual methods and other metareasoning approaches.en_US
dc.language.isoengen_US
dc.publisherNTNUen_US
dc.relation.ispartofseriesDoctoral theses at NTNU;2020:193
dc.relation.haspartPaper A: Houeland, Tor Gunnar Høst; Aamodt, Agnar. An Introspective Component-Based Approach for Meta-Level Reasoning in Clinical Decision Support Systems. I: Proceedings of the First Norwegian Artificial Intelligence Symposium. Tapir Akademisk Forlag 2009 ISBN 978-82-519-2519-8. s. 121-132en_US
dc.relation.haspartPaper B: Houeland, Tor Gunnar Høst; Aamodt, Agnar. The utility problem for lazy learners - towards a non-eager approach. Lecture Notes in Computer Science (LNCS) 2010 ;Volum 6176. s. 141-155 - The final authenticated version is available online at: https://doi.org/10.1007/978-3-642-14274-1_12en_US
dc.relation.haspartPaper C: Houeland, Tor Gunnar Høst. An Efficient Random Decision Tree Algorithm for Case-Based Reasoning Systems. I: Proceedings of the Twenty-Fourth International Florida Artificial Intelligence Research Society Conference. AAAI Press 2011 ISBN 978-1-57735-501-4. s. 401-406 https://aaai.org/ocs/index.php/FLAIRS/FLAIRS11/paper/view/2639en_US
dc.relation.haspartPaper D: Houeland, Tor Gunnar Høst; Aamodt, Agnar. An Efficient Hybrid Classification Algorithm - An Example from Palliative Care. Lecture Notes in Computer Science (LNCS) 2011 ;Volum 6679. s. 197-204 - The final authenticated version is available online at: https://doi.org/10.1007/978-3-642-21222-2_24en_US
dc.relation.haspartPaper E: Houeland, Tor Gunnar Høst; Bruland, Tore; Aamodt, Agnar; Langseth, Helge. Extended abstract: Combining CBR and BN using metareasoning. Published in: A. Kofod-Petersen et al. (Eds.) Eleventh Scandinavian Conference on Artificial Intelligence, pp. 189–190. IOS Press (2011) https://doi.org/10.3233/978-1-60750-754-3-189en_US
dc.relation.haspartPaper F: Houeland, Tor Gunnar Høst; Aamodt, Agnar. A learning system based on lazy metareasoning. Progress in Artificial Intelligence 2017 ;Volum 7.(2) s. 129-146 - The final authenticated version is available online at: https://doi.org/10.1007/s13748-017-0138-0en_US
dc.titleAutomated lazy metalearning in introspective reasoning systemsen_US
dc.typeDoctoral thesisen_US
dc.subject.nsiVDP::Technology: 500::Information and communication technology: 550en_US


Tilhørende fil(er)

Thumbnail

Denne innførselen finnes i følgende samling(er)

Vis enkel innførsel