antonior92 / attractors-and-smoothness-rnn Goto Github PK
View Code? Open in Web Editor NEWCompanion code to the 2020 AISTATS paper: ''Beyond exploding and vanishing gradients: analysing RNN training using attractors and smoothness"
Home Page: https://arxiv.org/pdf/1906.08482.pdf
License: MIT License