Automated rule generation for optimal data selection in robust neural network training
| dc.contributor.author | Verbytskyi O. S. | |
| dc.contributor.author | Gaidaienko O. V. | |
| dc.date.accessioned | 2026-06-09T09:06:00Z | |
| dc.date.issued | 2025-09-25 | |
| dc.description | Verbytskyi, O. S. Automated rule generation for optimal data selection in robust neural network training / O. S. Verbytskyi, O. V. Gaidaienko // Матеріали ХVІ міжнар. науково-технічна конф. "Інновації в суднобудуванні та океанотехніці". – Миколаїв : НУК. – 2025. – С. 631–633. | |
| dc.description.abstract1 | This study tackles the bottleneck of manual data curation by proposing a framework for automated data selection. We train a meta-model to quantify the "utility" of each data point using key metrics like entropy and representativeness. This model then generates human-readable rules to filter noisy datasets, creating optimized subsets for training. The goal is to simultaneously improve the final accuracy, robustness, and convergence speed of neural networks. | |
| dc.description.provenance | Submitted by Оксана Гайдаєнко (oksana.gaidaienko@nuos.edu.ua) on 2026-06-08T08:41:15Z workflow start=Step: reviewstep - action:claimaction No. of bitstreams: 1 Вербицкий О_4.pdf: 508286 bytes, checksum: dc3d1c1a54652f20585a222e27e260cf (MD5) | en |
| dc.description.provenance | Step: reviewstep - action:reviewaction Approved for entry into archive by Бондар Ольга (olga.bondar@nuos.edu.ua) on 2026-06-09T09:00:47Z (GMT) | en |
| dc.description.provenance | Step: editstep - action:editaction Approved for entry into archive by Бондар Ольга (olga.bondar@nuos.edu.ua) on 2026-06-09T09:05:28Z (GMT) | en |
| dc.description.provenance | Step: finaleditstep - action:finaleditaction Approved for entry into archive by Бондар Ольга (olga.bondar@nuos.edu.ua) on 2026-06-09T09:06:00Z (GMT) | en |
| dc.description.provenance | Made available in DSpace on 2026-06-09T09:06:00Z (GMT). No. of bitstreams: 1 Verbytskyi.pdf: 508286 bytes, checksum: dc3d1c1a54652f20585a222e27e260cf (MD5) Previous issue date: 2025-09-25 | en |
| dc.identifier.isbn | 978-966-321-487-0 | |
| dc.identifier.uri | https://eir.nuos.edu.ua/handle/123456789/13017 | |
| dc.language.iso | en | |
| dc.relation.ispartofseries | УДК ; 004.658.2:004.853 | |
| dc.subject | data selection | |
| dc.subject | data curation | |
| dc.subject | meta-learning | |
| dc.subject | active learning | |
| dc.subject | curriculum learning | |
| dc.subject | robust models | |
| dc.subject | neural networks. | |
| dc.title | Automated rule generation for optimal data selection in robust neural network training | |
| dc.type | Theses |
Файли
Контейнер файлів
1 - 1 з 1
Вантажиться...
- Назва:
- Verbytskyi.pdf
- Розмір:
- 496.37 KB
- Формат:
- Adobe Portable Document Format
Ліцензійна угода
1 - 1 з 1
Вантажиться...
- Назва:
- license.txt
- Розмір:
- 4.38 KB
- Формат:
- Item-specific license agreed upon to submission
- Опис: