Gates Are Not What You Need in RNNs

TL;DR

This paper introduces the Residual Recurrent Unit (RRU), a new RNN cell that omits gates, yet outperforms traditional gated cells like GRU and LSTM across various tasks, demonstrating robustness and simplicity.

Contribution

The paper proposes the RRU, a gate-free recurrent cell based on residual connections, linear transformations, ReLU, and normalization, outperforming traditional gated units.

Findings

RRU outperforms GRU and LSTM on multiple tasks

RRU shows better robustness to parameter tuning

RRU is simpler and does not require gating mechanisms

Abstract

Recurrent neural networks have flourished in many areas. Consequently, we can see new RNN cells being developed continuously, usually by creating or using gates in a new, original way. But what if we told you that gates in RNNs are redundant? In this paper, we propose a new recurrent cell called Residual Recurrent Unit (RRU) which beats traditional cells and does not employ a single gate. It is based on the residual shortcut connection, linear transformations, ReLU, and normalization. To evaluate our cell's effectiveness, we compare its performance against the widely-used GRU and LSTM cells and the recently proposed Mogrifier LSTM on several tasks including, polyphonic music modeling, language modeling, and sentiment analysis. Our experiments show that RRU outperforms the traditional gated units on most of these tasks. Also, it has better robustness to parameter selection, allowing immediate application in new tasks without much tuning. We have implemented the RRU in TensorFlow, and the code is made available at https://github.com/LUMII-Syslab/RRU .