Metaheuristics and machine learning: an approach with reinforcement learning assisting neural architecture search.
Saved in:
| Title: | Metaheuristics and machine learning: an approach with reinforcement learning assisting neural architecture search. |
|---|---|
| Authors: | Venske, Sandra Mara Scós1,2 (AUTHOR) ssvenske@unicentro.br, de Almeida, Carolina Paula2 (AUTHOR), Delgado, Myriam Regattieri1 (AUTHOR) |
| Source: | Journal of Heuristics. Aug2024, Vol. 30 Issue 3/4, p199-224. 26p. |
| Subjects: | Artificial neural networks, Machine learning, Reinforcement learning, Protein structure prediction, Genetic algorithms, Metaheuristic algorithms, Evolutionary algorithms |
| Abstract: | Methaheuristics (MHs) are techniques widely used for solving complex optimization problems. In recent years, the interest in combining MH and machine learning (ML) has grown. This integration can occur mainly in two ways: ML-in-MH and MH-in-ML. In the present work, we combine the techniques in both ways—ML-in-MH-in-ML, providing an approach in which ML is considered to improve the performance of an evolutionary algorithm (EA), whose solutions encode parameters of an ML model—artificial neural network (ANN). Our approach called TS in EA in ANN employs a reinforcement learning neighborhood (RLN) mutation based on Thompson sampling (TS). TS is a parameterless reinforcement learning method, used here to boost the EA performance. In the experiments, every candidate ANN solves a regression problem known as protein structure prediction deviation. We consider two protein datasets, one with 16,382 and the other with 45,730 samples. The results show that TS in EA in ANN performs significantly better than a canonical genetic algorithm (GA in ANN) and the evolutionary algorithm without reinforcement learning (EA in ANN). Analyses of the parameter's frequency are also performed comparing the approaches. Finally, comparisons with the literature show that except for one particular case in the largest dataset, TS in EA in ANN outperforms other approaches considered the state of the art for the addressed datasets. [ABSTRACT FROM AUTHOR] |
| Copyright of Journal of Heuristics is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| FullText | Links: – Type: pdflink Text: Availability: 1 |
|---|---|
| Header | DbId: egs DbLabel: Engineering Source An: 178402057 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: Metaheuristics and machine learning: an approach with reinforcement learning assisting neural architecture search. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Venske%2C+Sandra+Mara+Scós%22">Venske, Sandra Mara Scós</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<i> ssvenske@unicentro.br</i><br /><searchLink fieldCode="AR" term="%22de+Almeida%2C+Carolina+Paula%22">de Almeida, Carolina Paula</searchLink><relatesTo>2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Delgado%2C+Myriam+Regattieri%22">Delgado, Myriam Regattieri</searchLink><relatesTo>1</relatesTo> (AUTHOR) – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22Journal+of+Heuristics%22">Journal of Heuristics</searchLink>. Aug2024, Vol. 30 Issue 3/4, p199-224. 26p. – Name: Subject Label: Subjects Group: Su Data: <searchLink fieldCode="DE" term="%22Artificial+neural+networks%22">Artificial neural networks</searchLink><br /><searchLink fieldCode="DE" term="%22Machine+learning%22">Machine learning</searchLink><br /><searchLink fieldCode="DE" term="%22Reinforcement+learning%22">Reinforcement learning</searchLink><br /><searchLink fieldCode="DE" term="%22Protein+structure+prediction%22">Protein structure prediction</searchLink><br /><searchLink fieldCode="DE" term="%22Genetic+algorithms%22">Genetic algorithms</searchLink><br /><searchLink fieldCode="DE" term="%22Metaheuristic+algorithms%22">Metaheuristic algorithms</searchLink><br /><searchLink fieldCode="DE" term="%22Evolutionary+algorithms%22">Evolutionary algorithms</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: Methaheuristics (MHs) are techniques widely used for solving complex optimization problems. In recent years, the interest in combining MH and machine learning (ML) has grown. This integration can occur mainly in two ways: ML-in-MH and MH-in-ML. In the present work, we combine the techniques in both ways—ML-in-MH-in-ML, providing an approach in which ML is considered to improve the performance of an evolutionary algorithm (EA), whose solutions encode parameters of an ML model—artificial neural network (ANN). Our approach called TS in EA in ANN employs a reinforcement learning neighborhood (RLN) mutation based on Thompson sampling (TS). TS is a parameterless reinforcement learning method, used here to boost the EA performance. In the experiments, every candidate ANN solves a regression problem known as protein structure prediction deviation. We consider two protein datasets, one with 16,382 and the other with 45,730 samples. The results show that TS in EA in ANN performs significantly better than a canonical genetic algorithm (GA in ANN) and the evolutionary algorithm without reinforcement learning (EA in ANN). Analyses of the parameter's frequency are also performed comparing the approaches. Finally, comparisons with the literature show that except for one particular case in the largest dataset, TS in EA in ANN outperforms other approaches considered the state of the art for the addressed datasets. [ABSTRACT FROM AUTHOR] – Name: AbstractSuppliedCopyright Label: Group: Ab Data: <i>Copyright of Journal of Heuristics is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.) |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=178402057 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1007/s10732-024-09526-1 Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 26 StartPage: 199 Subjects: – SubjectFull: Artificial neural networks Type: general – SubjectFull: Machine learning Type: general – SubjectFull: Reinforcement learning Type: general – SubjectFull: Protein structure prediction Type: general – SubjectFull: Genetic algorithms Type: general – SubjectFull: Metaheuristic algorithms Type: general – SubjectFull: Evolutionary algorithms Type: general Titles: – TitleFull: Metaheuristics and machine learning: an approach with reinforcement learning assisting neural architecture search. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Venske, Sandra Mara Scós – PersonEntity: Name: NameFull: de Almeida, Carolina Paula – PersonEntity: Name: NameFull: Delgado, Myriam Regattieri IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 08 Text: Aug2024 Type: published Y: 2024 Identifiers: – Type: issn-print Value: 13811231 Numbering: – Type: volume Value: 30 – Type: issue Value: 3/4 Titles: – TitleFull: Journal of Heuristics Type: main |
| ResultId | 1 |
