Adversarial Dropout for Recurrent Neural Networks

TL;DR

This paper introduces adversarial dropout for RNNs, which improves generalization by adversarially disconnecting dominant neurons, leading to better performance on various sequential tasks.

Contribution

It proposes a novel adversarial dropout method for RNNs that enhances regularization by targeting influential neurons, outperforming standard dropout techniques.

Findings

Adversarial dropout improves RNN performance on sequential MNIST tasks.

The method enhances semi-supervised text classification accuracy.

It leads to better language modeling results.

Abstract

Successful application processing sequential data, such as text and speech, requires an improved generalization performance of recurrent neural networks (RNNs). Dropout techniques for RNNs were introduced to respond to these demands, but we conjecture that the dropout on RNNs could have been improved by adopting the adversarial concept. This paper investigates ways to improve the dropout for RNNs by utilizing intentionally generated dropout masks. Specifically, the guided dropout used in this research is called as adversarial dropout, which adversarially disconnects neurons that are dominantly used to predict correct targets over time. Our analysis showed that our regularizer, which consists of a gap between the original and the reconfigured RNNs, was the upper bound of the gap between the training and the inference phases of the random dropout. We demonstrated that minimizing our regularizer improved the effectiveness of the dropout for RNNs on sequential MNIST tasks, semi-supervised text classification tasks, and language modeling tasks.