Summary: | Recurrent neural networks (RNN) are efficient in modeling sequences for generation and classification, but their training is obstructed by the vanishing and exploding gradient issues. In this paper, we reformulate the RNN unit to learn the residual functions with reference to the hidden state instead of conventional gated mechanisms such as long short-term memory (LSTM) and the gated recurrent unit (GRU). The residual structure has two main highlights: firstly, it solves the gradient vanishing and exploding issues for large time-distributed scales; secondly, the residual structure promotes the optimizations for backward updates. In the experiments, we apply language modeling, emotion classification and polyphonic modeling to evaluate our layer compared with LSTM and GRU layers. The results show that our layer gives state-of-the-art performance, outperforms LSTM and GRU layers in terms of speed, and supports an accuracy competitive with that of the other methods.
|