A Model for Learning-Curve Estimation in Efficient Neural Architecture Search and Its Application in Predictive Health Maintenance
A persistent challenge in machine learning is the computational inefficiency of neural architecture search (NAS), particularly in resource-constrained domains like predictive maintenance. This work introduces a novel learning-curve estimation framework that reduces NAS computational costs by over 50...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-02-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/13/4/555 |
| _version_ | 1826582074737295360 |
|---|---|
| author | David Solís-Martín Juan Galán-Páez Joaquín Borrego-Díaz |
| author_facet | David Solís-Martín Juan Galán-Páez Joaquín Borrego-Díaz |
| author_sort | David Solís-Martín |
| collection | DOAJ |
| description | A persistent challenge in machine learning is the computational inefficiency of neural architecture search (NAS), particularly in resource-constrained domains like predictive maintenance. This work introduces a novel learning-curve estimation framework that reduces NAS computational costs by over 50% while maintaining model performance, addressing a critical bottleneck in automated machine learning design. By developing a data-driven estimator trained on 62 different predictive maintenance datasets, we demonstrate a generalized approach to early-stopping trials during neural network optimization. Our methodology not only reduces computational resources but also provides a transferable technique for efficient neural network architecture exploration across complex industrial monitoring tasks. The proposed approach achieves a remarkable balance between computational efficiency and model performance, with only a 2% performance degradation, showcasing a significant advancement in automated neural architecture optimization strategies. |
| first_indexed | 2025-03-14T15:00:11Z |
| format | Article |
| id | doaj.art-9bccabe4d9f246c0a42d9e0eb0c9915d |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2025-03-14T15:00:11Z |
| publishDate | 2025-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-9bccabe4d9f246c0a42d9e0eb0c9915d2025-02-25T13:36:44ZengMDPI AGMathematics2227-73902025-02-0113455510.3390/math13040555A Model for Learning-Curve Estimation in Efficient Neural Architecture Search and Its Application in Predictive Health MaintenanceDavid Solís-Martín0Juan Galán-Páez1Joaquín Borrego-Díaz2Department of Computer Science and Artificial Intelligence, Universidad de Sevilla, 41012 Sevilla, SpainDepartment of Computer Science and Artificial Intelligence, Universidad de Sevilla, 41012 Sevilla, SpainDepartment of Computer Science and Artificial Intelligence, Universidad de Sevilla, 41012 Sevilla, SpainA persistent challenge in machine learning is the computational inefficiency of neural architecture search (NAS), particularly in resource-constrained domains like predictive maintenance. This work introduces a novel learning-curve estimation framework that reduces NAS computational costs by over 50% while maintaining model performance, addressing a critical bottleneck in automated machine learning design. By developing a data-driven estimator trained on 62 different predictive maintenance datasets, we demonstrate a generalized approach to early-stopping trials during neural network optimization. Our methodology not only reduces computational resources but also provides a transferable technique for efficient neural network architecture exploration across complex industrial monitoring tasks. The proposed approach achieves a remarkable balance between computational efficiency and model performance, with only a 2% performance degradation, showcasing a significant advancement in automated neural architecture optimization strategies.https://www.mdpi.com/2227-7390/13/4/555learning curvesneural architecture searchpredictive maintenanceBayesian optimization |
| spellingShingle | David Solís-Martín Juan Galán-Páez Joaquín Borrego-Díaz A Model for Learning-Curve Estimation in Efficient Neural Architecture Search and Its Application in Predictive Health Maintenance Mathematics learning curves neural architecture search predictive maintenance Bayesian optimization |
| title | A Model for Learning-Curve Estimation in Efficient Neural Architecture Search and Its Application in Predictive Health Maintenance |
| title_full | A Model for Learning-Curve Estimation in Efficient Neural Architecture Search and Its Application in Predictive Health Maintenance |
| title_fullStr | A Model for Learning-Curve Estimation in Efficient Neural Architecture Search and Its Application in Predictive Health Maintenance |
| title_full_unstemmed | A Model for Learning-Curve Estimation in Efficient Neural Architecture Search and Its Application in Predictive Health Maintenance |
| title_short | A Model for Learning-Curve Estimation in Efficient Neural Architecture Search and Its Application in Predictive Health Maintenance |
| title_sort | model for learning curve estimation in efficient neural architecture search and its application in predictive health maintenance |
| topic | learning curves neural architecture search predictive maintenance Bayesian optimization |
| url | https://www.mdpi.com/2227-7390/13/4/555 |
| work_keys_str_mv | AT davidsolismartin amodelforlearningcurveestimationinefficientneuralarchitecturesearchanditsapplicationinpredictivehealthmaintenance AT juangalanpaez amodelforlearningcurveestimationinefficientneuralarchitecturesearchanditsapplicationinpredictivehealthmaintenance AT joaquinborregodiaz amodelforlearningcurveestimationinefficientneuralarchitecturesearchanditsapplicationinpredictivehealthmaintenance AT davidsolismartin modelforlearningcurveestimationinefficientneuralarchitecturesearchanditsapplicationinpredictivehealthmaintenance AT juangalanpaez modelforlearningcurveestimationinefficientneuralarchitecturesearchanditsapplicationinpredictivehealthmaintenance AT joaquinborregodiaz modelforlearningcurveestimationinefficientneuralarchitecturesearchanditsapplicationinpredictivehealthmaintenance |