Variable Computation in Recurrent Neural Networks

TL;DR

This paper introduces a modification to recurrent neural networks that enables them to adapt their computation at each step, improving efficiency and performance on sequential data tasks.

Contribution

It proposes a novel approach allowing RNNs to learn variable computation per step without prior sequence structure knowledge.

Findings

Models perform fewer operations while maintaining accuracy

Variable computation improves overall task performance

Enhanced efficiency over traditional fixed-computation RNNs

Abstract

Recurrent neural networks (RNNs) have been used extensively and with increasing success to model various types of sequential data. Much of this progress has been achieved through devising recurrent units and architectures with the flexibility to capture complex statistics in the data, such as long range dependency or localized attention phenomena. However, while many sequential data (such as video, speech or language) can have highly variable information flow, most recurrent models still consume input features at a constant rate and perform a constant number of computations per time step, which can be detrimental to both speed and model capacity. In this paper, we explore a modification to existing recurrent units which allows them to learn to vary the amount of computation they perform at each step, without prior knowledge of the sequence's time structure. We show experimentally that not only do our models require fewer operations, they also lead to better performance overall on evaluation tasks.