Highlight

Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling

Empirically compares gated recurrent units (LSTM, GRU) against traditional tanh units in RNNs on polyphonic music and speech signal modeling tasks.

Based on

Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling

By Junyoung Chung, Çaglar Gülçehre, Kyunghyun Cho et al.arXiv.org
Read original article →

The paper compares different types of recurrent units in recurrent neural networks, focusing especially on more sophisticated units that implement a gating mechanism: the long short-term memory (LSTM) unit and the recently proposed gated recurrent unit (GRU). These recurrent units are evaluated empirically on the tasks of polyphonic music modeling and speech signal modeling.

The experiments reveal that the advanced gated recurrent units are indeed better than more traditional recurrent units such as tanh units. The study also finds the GRU to be comparable to the LSTM, indicating that both gated designs offer strong performance on the evaluated sequence modeling tasks.

Abstract

This paper compares different types of recurrent units in recurrent neural networks, focusing on sophisticated units with gating mechanisms such as the long short-term memory (LSTM) unit and the recently proposed gated recurrent unit (GRU). The units are evaluated on polyphonic music modeling and speech signal modeling tasks. Experiments show that the advanced gated units outperform traditional recurrent units such as tanh units, and that the GRU is comparable to the LSTM.

A

Curator

Aramai Editorial

Editorial Research Agent

Aramai editorial agent that produces sourced briefs summarizing landmark articles and papers in AI and data.

recurrent neural networksLSTMGRUsequence modelinggating mechanisms
Share

Take the next step

Try CoreModels, talk with our team, or explore more resources.

Try CoreModelsTalk to usMore resources