Skip to Main content Skip to Navigation
Conference papers

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

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.
Document type :
Conference papers
Complete list of metadata

Cited literature [30 references]  Display  Hide  Download

https://hal.inria.fr/hal-01821037
Contributor : Hal Ifip <>
Submitted on : Friday, June 22, 2018 - 11:44:24 AM
Last modification on : Friday, June 22, 2018 - 12:00:55 PM
Long-term archiving on: : Monday, September 24, 2018 - 7:27:28 PM

File

467708_1_En_15_Chapter.pdf
Files produced by the author(s)

Licence


Distributed under a Creative Commons Attribution 4.0 International License

Identifiers

Citation

Adrian 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⟩

Share

Metrics

Record views

186

Files downloads

13