Scalable reinforcement learning-based neural architecture search
Neural Computing and Applications · 2025-01 · 37 (1) · 231-261
DOI: 10.1007/s00521-024-10445-2 ↗Kurz gefasst
Fachartikel in der Zeitschrift Neural Computing and Applications, online erschienen am 18.11.2024. Verfasst von Amber Cassimon und 2 weiteren Autor:innen. Die Arbeit wurde bisher 21-mal zitiert (Crossref) und nennt 48 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Fonds Wetenschappelijk Onderzoek, Vlaamse Overheid.
Abstract (Originalsprache)
We assess the feasibility of a reusable neural architecture search agent aimed at amortizing the initial time-investment in building a good search strategy. We do this through the use of Reinforcement Learning, where an agent learns to iteratively select the best way to modify a given neural network architecture. This is achieved using a transformer-based agent design trained using the Ape-X algorithm. We consider both the NAS-Bench-101 and NAS-Bench-301 settings, and compare against various known strong baselines, such as local search and random search. While achieving competitive performance on both benchmarks, the amount of training required for the much larger NAS-Bench-301 is only marginally greater than NAS-Bench-101, illustrating the strong scaling properties of our agent. Our agent is able to achieve strong performance, but the choice of values for certain parameters are crucial to ensuring the succesful training of the agent. We provide some guidance for the selection of appropriate values for hyperparameters through a detailed description of our experimental setup and several ablation studies.
Förderung
- Fonds Wetenschappelijk Onderzoek
- Vlaamse Overheid
Zitieren
APA
Cassimon, A., Mercelis, S., & Mets, K. (2025). Scalable reinforcement learning-based neural architecture search. Neural Computing and Applications, 37(1), 231-261. https://doi.org/10.1007/s00521-024-10445-2
Vancouver
Cassimon A, Mercelis S, Mets K. Scalable reinforcement learning-based neural architecture search. Neural Computing and Applications. 2025;37(1):231-261. doi:10.1007/s00521-024-10445-2
BibTeX
@article{Cassimon2025,
author = {Cassimon, Amber and Mercelis, Siegfried and Mets, Kevin},
title = {Scalable reinforcement learning-based neural architecture search},
journal = {Neural Computing and Applications},
year = {2025},
volume = {37},
number = {1},
pages = {231--261},
doi = {10.1007/s00521-024-10445-2}
}
EUVIMED