Recurrent Neural Networks with Mixed Hierarchical Structures and EM Algorithm for Natural Language Processing

TL;DR

This paper introduces EM-HRNN, a novel RNN architecture with a latent indicator layer and EM algorithm to capture hierarchical language structures, improving document classification without pre-training.

Contribution

The paper proposes a new EM-based RNN model with a latent indicator layer for implicit hierarchical structure learning in language processing.

Findings

EM-HRNN outperforms other RNN models in document classification

Model achieves comparable performance to BERT without pre-training

Bootstrap strategies improve training efficiency on long texts

Abstract

How to obtain hierarchical representations with an increasing level of abstraction becomes one of the key issues of learning with deep neural networks. A variety of RNN models have recently been proposed to incorporate both explicit and implicit hierarchical information in modeling languages in the literature. In this paper, we propose a novel approach called the latent indicator layer to identify and learn implicit hierarchical information (e.g., phrases), and further develop an EM algorithm to handle the latent indicator layer in training. The latent indicator layer further simplifies a text's hierarchical structure, which allows us to seamlessly integrate different levels of attention mechanisms into the structure. We called the resulting architecture as the EM-HRNN model. Furthermore, we develop two bootstrap strategies to effectively and efficiently train the EM-HRNN model on long text documents. Simulation studies and real data applications demonstrate that the EM-HRNN model with bootstrap training outperforms other RNN-based models in document classification tasks. The performance of the EM-HRNN model is comparable to a Transformer-based method called Bert-base, though the former is much smaller model and does not require pre-training.