Online Representation Learning in Recurrent Neural Language Models

TL;DR

This paper introduces an online learning extension for recurrent neural network language models that adaptively updates representations to improve accuracy and efficiency.

Contribution

It presents a novel online learning method that maintains and updates separate text unit vectors, reducing parameters and computational costs.

Findings

Improved language modeling accuracy

Reduced model parameters

Lower computational requirements

Abstract

We investigate an extension of continuous online learning in recurrent neural network language models. The model keeps a separate vector representation of the current unit of text being processed and adaptively adjusts it after each prediction. The initial experiments give promising results, indicating that the method is able to increase language modelling accuracy, while also decreasing the parameters needed to store the model along with the computation required at each step.