most citedBeyond exploding and vanishing gradients: analysing RNN training using attractors and smoothness

50 citations · 50 across the 1 of their papers we have counts for

collaborators

1 paper

cs.LG2019★ 50 cited

Beyond exploding and vanishing gradients: analysing RNN training using attractors and smoothness

Antônio H. Ribeiro, Koen Tiels, Luis A. Aguirre +1

The exploding and vanishing gradient problem has been the major conceptual principle behind most architecture and training improvements in recurrent neural networks (RNNs) during t…