Skip to Main content Skip to Navigation
Conference papers

Unsupervised Layer-Wise Model Selection in Deep Neural Networks

Ludovic Arnold 1, 2 Hélène Paugam-Moisy 3, 2 Michèle Sebag 1, 2
2 TAO - Machine Learning and Optimisation
CNRS - Centre National de la Recherche Scientifique : UMR8623, Inria Saclay - Ile de France, UP11 - Université Paris-Sud - Paris 11, LRI - Laboratoire de Recherche en Informatique
3 DM2L - Data Mining and Machine Learning
LIRIS - Laboratoire d'InfoRmatique en Image et Systèmes d'information
Abstract : Deep Neural Networks (DNN) propose a new and efficient ML architecture based on the layer-wise building of several representation layers. A critical issue for DNNs remains model selection, e.g. selecting the number of neurons in each DNN layer. The hyper-parameter search space exponentially increases with the number of layers, making the popular grid search-based approach used for finding good hyper-parameter values intractable. The question investigated in this paper is whether the unsupervised, layer-wise methodology used to train a DNN can be extended to model selection as well. The proposed approach, considering an unsupervised criterion, empirically examines whether model selection is a modular optimization problem, and can be tackled in a layer-wise manner. Preliminary results on the MNIST data set suggest the answer is positive. Further, some unexpected results regarding the optimal size of layers depending on the training process, are reported and discussed.
Document type :
Conference papers
Complete list of metadata

Cited literature [19 references]  Display  Hide  Download
Contributor : Ludovic Arnold <>
Submitted on : Tuesday, June 1, 2010 - 5:34:08 PM
Last modification on : Tuesday, December 8, 2020 - 10:30:16 AM
Long-term archiving on: : Friday, October 19, 2012 - 3:30:30 PM


Files produced by the author(s)



Ludovic Arnold, Hélène Paugam-Moisy, Michèle Sebag. Unsupervised Layer-Wise Model Selection in Deep Neural Networks. 19th European Conference on Artificial Intelligence (ECAI'10), Aug 2010, Lisbon, Portugal. ⟨10.3233/978-1-60750-606-5-915⟩. ⟨hal-00488338⟩



Record views


Files downloads