An Empirical Exploration of Recurrent Network Architectures

It is found that adding a bias of 1 to the LSTM's forget gate closes the gap between the L STM and the recently-introduced Gated Recurrent Unit (GRU) on some but not all tasks.