Theano: Deep Learning on GPUs with Python
Departement d’Informatique et Recherche Operationnelle, Universite de Montreal, 2920 Chemin de la tour, Montreal, Quebec, Canada, H3T 1J8
Journal of Machine Learning Research 1, 1-48, 2011
@article{bergstra2011theano,
title={Theano: Deep Learning on GPUs with Python},
author={Bergstra, J. and Bastien, F. and Breuleux, O. and Lamblin, P. and Pascanu, R. and Delalleau, O. and Desjardins, G. and Warde-Farley, D. and Goodfellow, I. and Bergeron, A. and others},
year={2011}
}
In this paper, we present Theano, a framework in the Python programming language for defining, optimizing and evaluating expressions involving high-level operations on tensors. Theano offers most of NumPy’s functionality, but adds automatic symbolic differentiation, GPU support, and faster expression evaluation. Theano is a general mathematical tool, but it was developed with the goal of facilitating research in deep learning. The Deep Learning Tutorials introduce recent advances in deep learning, and showcase how Theano makes such algorithms compact, elegant, and fast.
January 16, 2012 by hgpu
