A Dynamic Early Stopping Criterion for Random Search in SVM Hyperparameter Optimization - Archive ouverte HAL Access content directly
Conference Papers Year : 2018

A Dynamic Early Stopping Criterion for Random Search in SVM Hyperparameter Optimization

(1) , (2)
1
2
Adrian Cătălin Florea
  • Function : Author
  • PersonId : 1033467
Răzvan Andonie
  • Function : Author
  • PersonId : 872190

Abstract

We introduce a dynamic early stopping condition for Random Search optimization algorithms. We test our algorithm for SVM hyperparameter optimization for classification tasks, on six commonly used datasets. According to the experimental results, we reduce significantly the number of trials used. Since each trial requires a re-training of the SVM model, our method accelerates the RS optimization. The code runs on a multi-core system and we analyze the achieved scalability for an increasing number of cores.
Fichier principal
Vignette du fichier
467708_1_En_15_Chapter.pdf (499.49 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01821037 , version 1 (22-06-2018)

Licence

Attribution - CC BY 4.0

Identifiers

Cite

Adrian Cătălin Florea, Răzvan Andonie. A Dynamic Early Stopping Criterion for Random Search in SVM Hyperparameter Optimization. 14th IFIP International Conference on Artificial Intelligence Applications and Innovations (AIAI), May 2018, Rhodes, Greece. pp.168-180, ⟨10.1007/978-3-319-92007-8_15⟩. ⟨hal-01821037⟩
120 View
73 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More