AI summary

This paper compares LSTM, GRU, and tanh units on polyphonic music datasets and speech signal modeling tasks; the GRU and LSTM units outperform the traditional tanh unit, with GRU being better than LSTM on some datasets, suggesting that the choice of the type of gated recurrent unit may depend heavily on the dataset and corresponding task.

Abstract

In this paper we compare different types of recurrent units in recurrent neural networks (RNNs). Especially, we focus on more sophisticated units that implement a gating mechanism, such as a long short-term memory (LSTM) unit and a recently proposed gated recurrent unit (GRU). We evaluate these recurrent units on the tasks of polyphonic music modeling and speech signal modeling. Our experiments revealed that these advanced recurrent units are indeed better than more traditional recurrent units such as tanh units. Also, we found GRU to be comparable to LSTM.