Giancarlo Kerg
Cited by
Cited by
Non-normal Recurrent Neural Network (nnRNN): learning long time dependencies while improving expressivity with transient dynamics
G Kerg, K Goyette, MP Touzel, G Gidel, E Vorontsov, Y Bengio, G Lajoie
arXiv preprint arXiv:1905.12080, 2019
h-detach: Modifying the LSTM gradient towards better optimization
B Kanuparthi, D Arpit, G Kerg, NR Ke, I Mitliagkas, Y Bengio
International Conference on Learning Representations, 2018
Safe screening for support vector machines
J Zimmert, CS de Witt, G Kerg, M Kloft
NIPS 2015 Workshop on Optimization in Machine Learning (OPT), 2015
Untangling tradeoffs between recurrence and self-attention in artificial neural networks
G Kerg, B Kanuparthi, AG ALIAS PARTH GOYAL, K Goyette, Y Bengio, ...
Advances in Neural Information Processing Systems 33, 2020
Catastrophic Fisher Explosion: Early Phase Fisher Matrix Impacts Generalization
S Jastrzebski, D Arpit, O Astrand, G Kerg, H Wang, C Xiong, R Socher, ...
arXiv preprint arXiv:2012.14193, 2020
Advantages of biologically-inspired adaptive neural activation in RNNs during learning
V Geadah, G Kerg, S Horoi, G Wolf, G Lajoie
arXiv preprint arXiv:2006.12253, 2020
The system can't perform the operation now. Try again later.
Articles 1–6