Understanding Hidden Memories of Recurrent Neural Networks

Abstract

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.

arXiv commentPublished at IEEE Conference on Visual Analytics Science and Technology (IEEE VAST 2017)

References in corpus (9)

• Neural Machine Translation by Jointly Learning to Align and Translate
• Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
• LSTM: A Search Space Odyssey
• On the Properties of Neural Machine Translation: Encoder-Decoder Approaches
• Visualizing and Understanding Recurrent Networks
• Automatic Rule Extraction from Long Short Term Memory Networks
• Memory Visualization for Gated Recurrent Neural Networks in Speech Recognition

Cited by in corpus (8)

• Evaluating the Readability of Force Directed Graph Layouts: A Deep Learning Approach
• DRLViz: Understanding Decisions and Memory in Deep Reinforcement Learning
• A Survey on Understanding, Visualizations, and Explanation of Deep Neural Networks
• RuleMatrix: Visualizing and Understanding Classifiers with Rules
• HypoML: Visual Analysis for Hypothesis-based Evaluation of Machine Learning Models
• LAMVI-2: A Visual Tool for Comparing and Tuning Word Embedding Models
• Recurrent Neural Networks based Obesity Status Prediction Using Activity Data
• Predicting protein secondary structure with Neural Machine Translation