Toward Optimal Run Racing: Application to Deep Learning Calibration - Archive ouverte HAL Access content directly
Preprints, Working Papers, ... Year :

Toward Optimal Run Racing: Application to Deep Learning Calibration

(1) , (1) , (1) , (2, 3) , (3, 2) , (1) , (1)
1
2
3
Olivier Bousquet
Sylvain Gelly
  • Function : Author
Karol Kurach
  • Function : Author
Olivier Teytaud
Damien Vincent
  • Function : Author

Abstract

This paper aims at one-shot learning of deep neural nets, where a highly parallel setting is considered to address the algorithm calibration problem - selecting the best neural architecture and learning hyper-parameter values depending on the dataset at hand. The notoriously expensive calibration problem is optimally reduced by detecting and early stopping non-optimal runs. The theoretical contribution regards the optimality guarantees within the multiple hypothesis testing framework. Experimentations on the Cifar10, PTB and Wiki benchmarks demonstrate the relevance of the approach with a principled and consistent improvement on the state of the art with no extra hyper-parameter.

Dates and versions

hal-01634381 , version 1 (14-11-2017)

Identifiers

Cite

Olivier Bousquet, Sylvain Gelly, Karol Kurach, Marc Schoenauer, Michèle Sebag, et al.. Toward Optimal Run Racing: Application to Deep Learning Calibration. 2017. ⟨hal-01634381⟩
137 View
0 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More