Razvan Pascanu
Razvan Pascanu
Research Scientist at Google DeepMind
Verifierad e-postadress på google.com - Startsida
Titel
Citeras av
Citeras av
År
On the difficulty of training recurrent neural networks
R Pascanu, T Mikolov, Y Bengio
International conference on machine learning, 1310-1318, 2013
36602013
Theano: a CPU and GPU math expression compiler
J Bergstra, O Breuleux, F Bastien, P Lamblin, R Pascanu, G Desjardins, ...
Proceedings of the Python for scientific computing conference (SciPy) 4 (3), 1-7, 2010
18362010
Overcoming catastrophic forgetting in neural networks
J Kirkpatrick, R Pascanu, N Rabinowitz, J Veness, G Desjardins, AA Rusu, ...
Proceedings of the national academy of sciences 114 (13), 3521-3526, 2017
16322017
Theano: new features and speed improvements
F Bastien, P Lamblin, R Pascanu, J Bergstra, I Goodfellow, A Bergeron, ...
arXiv preprint arXiv:1211.5590, 2012
15202012
Identifying and attacking the saddle point problem in high-dimensional non-convex optimization
Y Dauphin, R Pascanu, C Gulcehre, K Cho, S Ganguli, Y Bengio
arXiv preprint arXiv:1406.2572, 2014
10862014
Relational inductive biases, deep learning, and graph networks
PW Battaglia, JB Hamrick, V Bapst, A Sanchez-Gonzalez, V Zambaldi, ...
arXiv preprint arXiv:1806.01261, 2018
9962018
A simple neural network module for relational reasoning
A Santoro, D Raposo, DGT Barrett, M Malinowski, R Pascanu, P Battaglia, ...
arXiv preprint arXiv:1706.01427, 2017
9172017
Progressive neural networks
AA Rusu, NC Rabinowitz, G Desjardins, H Soyer, J Kirkpatrick, ...
arXiv preprint arXiv:1606.04671, 2016
9152016
On the number of linear regions of deep neural networks
G Montúfar, R Pascanu, K Cho, Y Bengio
arXiv preprint arXiv:1402.1869, 2014
8032014
How to construct deep recurrent neural networks
R Pascanu, C Gulcehre, K Cho, Y Bengio
arXiv preprint arXiv:1312.6026, 2013
7842013
Theano: A CPU and GPU math compiler in Python
J Bergstra, O Breuleux, F Bastien, P Lamblin, R Pascanu, G Desjardins, ...
Proc. 9th Python in Science Conf 1, 3-10, 2010
7092010
Interaction networks for learning about objects, relations and physics
PW Battaglia, R Pascanu, M Lai, D Rezende, K Kavukcuoglu
arXiv preprint arXiv:1612.00222, 2016
6622016
Theano: A Python framework for fast computation of mathematical expressions
R Al-Rfou, G Alain, A Almahairi, C Angermueller, D Bahdanau, N Ballas, ...
arXiv e-prints, arXiv: 1605.02688, 2016
6072016
Learning to navigate in complex environments
P Mirowski, R Pascanu, F Viola, H Soyer, AJ Ballard, A Banino, M Denil, ...
arXiv preprint arXiv:1611.03673, 2016
5252016
Advances in optimizing recurrent networks
Y Bengio, N Boulanger-Lewandowski, R Pascanu
2013 IEEE International Conference on Acoustics, Speech and Signal …, 2013
4872013
Understanding the exploding gradient problem
R Pascanu, T Mikolov, Y Bengio
CoRR, abs/1211.5063 2 (417), 1, 2012
4502012
Meta-learning with latent embedding optimization
AA Rusu, D Rao, J Sygnowski, O Vinyals, R Pascanu, S Osindero, ...
arXiv preprint arXiv:1807.05960, 2018
3772018
Imagination-augmented agents for deep reinforcement learning
S Racanière, T Weber, DP Reichert, L Buesing, A Guez, D Rezende, ...
Proceedings of the 31st International Conference on Neural Information …, 2017
360*2017
Pylearn2: a machine learning research library
IJ Goodfellow, D Warde-Farley, P Lamblin, V Dumoulin, M Mirza, ...
arXiv preprint arXiv:1308.4214, 2013
3322013
Sim-to-real robot learning from pixels with progressive nets
AA Rusu, M Večerík, T Rothörl, N Heess, R Pascanu, R Hadsell
Conference on Robot Learning, 262-270, 2017
3232017
Systemet kan inte utföra åtgärden just nu. Försök igen senare.
Artiklar 1–20