Understanding Hidden Memories of Recurrent Neural Networks

Recurrent neural networks (RNNs) have been successfully applied to various natural language processing (NLP) tasks and achieved better results than conventional methods. However, the lack of understanding of the mechanisms behind their effectiveness limits further improvements on their architectures. In this paper, we present a visual analytics method for understanding and comparing RNN models for NLP tasks. We propose a technique to explain the function of individual hidden state units based on their expected response to input texts. We then co-cluster hidden state units and words based on the expected response and visualize co-clustering results as memory chips and word clouds to provide more structured knowledge on RNNs’ hidden states. We also propose a glyph-based sequence visualization based on aggregate information to analyze the behavior of an RNN’s hidden state at the sentence-level. The usability and effectiveness of our method are demonstrated through case studies and reviews from domain experts.

1 Introduction

The interface of RNNVis. The control panel (A) shows parameter settings of an RNN and allows users to adjust visualization style. The main view (B-D) contains glyph-based sentence visualization (B), memory chips visualization for hidden state clusters (C), and word clouds visualization for word clusters (D). The detail view (E) shows the distributions of models responses to selected word “of” and interpretations of selected hidden units.