Author Identification using Multi-headed Recurrent Neural Networks

TL;DR

This paper introduces a multi-headed RNN approach for author identification, sharing a recurrent layer among authors while using separate output heads, achieving competitive results across multiple languages.

Contribution

It presents a novel multi-headed RNN architecture that effectively models individual author styles with limited data, outperforming existing methods in some languages.

Findings

Ranked first in two of four languages

Effective modeling of author style with shared recurrent layer

Competitive performance in author identification

Abstract

Recurrent neural networks (RNNs) are very good at modelling the flow of text, but typically need to be trained on a far larger corpus than is available for the PAN 2015 Author Identification task. This paper describes a novel approach where the output layer of a character-level RNN language model is split into several independent predictive sub-models, each representing an author, while the recurrent layer is shared by all. This allows the recurrent layer to model the language as a whole without over-fitting, while the outputs select aspects of the underlying model that reflect their author's style. The method proves competitive, ranking first in two of the four languages.