How to Construct Deep Recurrent Neural Networks

TL;DR

This paper investigates how to deepen recurrent neural networks by analyzing their architecture, proposing novel deep RNN structures, and demonstrating their improved performance on music prediction and language modeling tasks.

Contribution

The paper introduces two new deep RNN architectures, distinct from stacking layers, with an innovative interpretation using neural operators.

Findings

Deep RNNs outperform shallow RNNs in experiments.

Proposed architectures benefit from increased depth.

Analysis clarifies the concept of depth in RNNs.

Abstract

In this paper, we explore different ways to extend a recurrent neural network (RNN) to a \textit{deep} RNN. We start by arguing that the concept of depth in an RNN is not as clear as it is in feedforward neural networks. By carefully analyzing and understanding the architecture of an RNN, however, we find three points of an RNN which may be made deeper; (1) input-to-hidden function, (2) hidden-to-hidden transition and (3) hidden-to-output function. Based on this observation, we propose two novel architectures of a deep RNN which are orthogonal to an earlier attempt of stacking multiple recurrent layers to build a deep RNN (Schmidhuber, 1992; El Hihi and Bengio, 1996). We provide an alternative interpretation of these deep RNNs using a novel framework based on neural operators. The proposed deep RNNs are empirically evaluated on the tasks of polyphonic music prediction and language modeling. The experimental result supports our claim that the proposed deep RNNs benefit from the depth and outperform the conventional, shallow RNNs.