A Hierarchical Latent Variable Encoder-Decoder Model for Generating Dialogues

Authors

  • Iulian Serban University of Montreal
  • Alessandro Sordoni Maluuba Inc
  • Ryan Lowe McGill University
  • Laurent Charlin HEC Montréal
  • Joelle Pineau McGill University
  • Aaron Courville University of Montreal
  • Yoshua Bengio University of Montreal

DOI:

https://doi.org/10.1609/aaai.v31i1.10983

Keywords:

Dialogue System, Conversational System, Chatbot, Neural Network, Deep Learning, Generative Models, Variational Autoencoder, Latent Variable Model, Variational Learning, Twitter

Abstract

Sequential data often possesses hierarchical structures with complex dependencies between sub-sequences, such as found between the utterances in a dialogue. To model these dependencies in a generative framework, we propose a neural network-based generative architecture, with stochastic latent variables that span a variable number of time steps. We apply the proposed model to the task of dialogue response generation and compare it with other recent neural-network architectures. We evaluate the model performance through a human evaluation study. The experiments demonstrate that our model improves upon recently proposed models and that the latent variables facilitate both the generation of meaningful, long and diverse responses and maintaining dialogue state.

Downloads

Published

2017-02-12

How to Cite

Serban, I., Sordoni, A., Lowe, R., Charlin, L., Pineau, J., Courville, A., & Bengio, Y. (2017). A Hierarchical Latent Variable Encoder-Decoder Model for Generating Dialogues. Proceedings of the AAAI Conference on Artificial Intelligence, 31(1). https://doi.org/10.1609/aaai.v31i1.10983