Greedy layer-wise training of deep networks,” (2007)

by Y Bengio, P Lamblin, D Popovici, H Larochelle
Venue:in Neural Information Processing Systems (NIPS),