Logo LVSN
EnglishAccueil
A proposPersonnesRecherchePublicationsEvenementsProfil
A propos
Séminaires


Les Séminaires CerVIM, Université Laval ont lieu le vendredi à 11h00.
Veuillez consulter le programme pour plus de détails.

 

 

 

 

CERVIM

REPARTI

MIVIM

04-03-2010

Détails: tutoriel (4-5 mars 2010) / Tutorial Details (March 4-5, 2010)

Hugo Larochelle
Dept. of Computer Science, University of Toronto


Littérature : Réseaux de neurones profonds (suite)



Publications on Deep Neural Networks (suite)

Convolutional deep belief networks for scalable unsupervised learning of hierarchical representations
http://www.stanford.edu/~hllee/icml09-ConvolutionalDeepBeliefNetworks.pdf

Unsupervised feature learning for audio classification using convolutional deep belief networks
http://www.stanford.edu/~hllee/nips09-AudioConvolutionalDBN.pdf

A Unified Architecture for Natural Language Processing: Deep Neural Networks with Multitask Learning
http://ronan.collobert.com/pub/matos/2008_nlp_icml.pdf

Why Does Unsupervised Pre-training Help Deep Learning?
http://www-etud.iro.umontreal.ca/~erhandum/files/publications/erhan10a.pdf

Reinforcement Learning with Factored States and Actions
http://jmlr.csail.mit.edu/papers/volume5/sallans04a/sallans04a.pdf

Factored Conditional Restricted Boltzmann Machines for Modeling Motion Style
http://cs.nyu.edu/~gwtaylor/publications/icml2009/fcrbm_icml.pdf
DEMO: http://www.cs.nyu.edu/~gwtaylor/publications/icml2009/

Efficient Learning of Deep Boltzmann Machines
http://www.cs.toronto.edu/~larocheh/publications/aistats_2010_dbm_recnet.pdf





     
   
   

©2002-. Laboratoire de Vision et Systèmes Numériques. Tous droits réservés