Elamur Research⌕
☰
home / papers / 2017 / A Network-based End-to-End Trainable Task-oriented Dialogue System

A Network-based End-to-End Trainable Task-oriented Dialogue System

Paper · 2017 · Tsung-Hsien Wen, David Vandyke, Nikola Mrkšić et al.

Summary

The paper addresses the construction of dialogue systems that help a user accomplish a specific task, such as finding a restaurant. Such systems usually consist of many separate modules, each of which requires either manual design or costly annotation. The authors propose a neural network-based system that takes text as input and returns text. It can be trained end-to-end, yet it retains a modular structure. An intent network based on an LSTM or CNN encodes the user's utterance. Belief trackers estimate a probability distribution over slot–value pairs. Based on these estimates, the system queries the database. A policy network combines all these signals into a single action vector. From this vector, the generator produces a delexicalised response, into which actual values from the database are then substituted. The training data were collected via crowdsourcing using a novel pipelined Wizard-of-Oz scheme that directly yields annotations for the trackers. Training on a few hundred dialogues in the restaurant search domain produced a system that human judges rated higher than a strong modular baseline with handcrafted rules. The work shows that end-to-end training is practical for task-oriented dialogue even with a small amount of data.

Key points

  • Architecture: an intent network, belief trackers (a Jordan-type RNN with a CNN feature extractor and shared weights), a DB operator, a policy network and an LSTM generator. Delexicalisation replaces values with generic tokens, such as <v.food>, and thus reduces the need for data.
  • Attention and weighted decoding: an attention mechanism over the tracker states increases task success by approximately 5%, while BLEU rises by only about 0.01. Weighted decoding yields a further improvement.
  • Human evaluation: on Amazon Mechanical Turk, the full model was tested in 245 dialogues. Subjective success was 98%, comprehension 4.11 out of 5, and naturalness 4.05 out of 5.
  • Comparison with HDC (164 comparisons): success of 96.95% versus 95.12%, with an average of 3.95 turns versus 4.54. The neural system was preferred by 50% of the judges, HDC by 24.39% (p < 0.005). It also won on naturalness and comprehension.
  • Limitations: the system works with text only, does not handle noisy speech recognition output, and does not ask the user for clarification when it is uncertain. Whether the approach scales to broader domains remains an open question.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓University of CambridgeCambridge · United Kingdom243 works · 121,396 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 works

🧭 Topics 1Research subjects and languages the work is about.Dialogue systems · 624 works

💡 Concepts 1Key ideas and methods explained on our concept pages.Chatbots · 54 works

Cite

Cite as Tsung-Hsien Wen, David Vandyke, Nikola Mrkšić, Milica Gašić, Lina M. Rojas-Barahona, Pei-Hao Su, Stefan Ultes and Steve Young. 2017. A Network-based End-to-End Trainable Task-oriented Dialogue System. Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers.

The paper itself, not this page. BibTeX from the library's record of the work. arXiv 1604.04562

@inproceedings{wen2017network,
  title = {{A Network-based End-to-End Trainable Task-oriented Dialogue System}},
  author = {Tsung-Hsien Wen, David Vandyke, Nikola Mrkšić, Milica Gašić, Lina M. Rojas-Barahona, Pei-Hao Su, Stefan Ultes and Steve Young},
  booktitle = {Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers},
  year = {2017},
  eprint = {1604.04562},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/1604.04562}
}

References

38 references, 16 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Dzmitry Bahdanau, Kyunghyun Cho, and Yoshua Bengio. 2014. Neural machine translation by jointly learning to align and translate. arXiv preprint:1409.0473.
  2. Jonathan Berant, Andrew Chou, Roy Frostig, and Percy Liang. 2013. Semantic parsing on Freebase from question-answer pairs. In EMNLP, pages 1533–1544, Seattle, Washington, USA. ACL.
  3. Dan Bohus and Alexander I. Rudnicky, 2008. Sorry, I Didn’t Catch That!, pages 123–154. Springer Netherlands, Dordrecht. find
  4. Kyunghyun Cho, Bart van Merrienboer, Caglar Gulcehre, Dzmitry Bahdanau, Fethi Bougares, Holger Schwenk, and Yoshua Bengio. 2014. Learning phrase representations using rnn encoder–decoder for statistical machine translation. In EMNLP, pages 1724–1734, Doha, Qatar, October. ACL.
  5. Laurens Van der Maaten and Geoffrey Hinton. 2008. Visualizing Data using t-SNE. JMLR. find
  6. Milica Gašić, Catherine Breslin, Matthew Henderson, Dongho Kim, Martin Szummer, Blaise Thomson, Pirros Tsiakoulis, and Steve Young. 2013. On-line policy optimisation of bayesian spoken dialogue systems via human interaction. In ICASSP, pages 8367–8371, May. find
  7. Matthew Henderson, Blaise Thomson, and Steve Young. 2014. Word-based dialog state tracking with recurrent neural networks. In SIGDIAL, pages 292–299, Philadelphia, PA, USA, June. ACL.
  8. Matthew Henderson. 2015. Machine learning for dialog state tracking: A review. In Machine Learning in Spoken Language Processing Workshop. find
  9. Karl Moritz Hermann, Tomás Kociský, Edward Grefenstette, Lasse Espeholt, Will Kay, Mustafa Suleyman, and Phil Blunsom. 2015. Teaching machines to read and comprehend. In NIPS, pages 1693–1701, Montreal, Canada. MIT Press.
  10. Sepp Hochreiter and Jürgen Schmidhuber. 1997. Long short-term memory. Neural Compututation, 9(8):1735–1780, November. find
  11. Michael I. Jordan. 1989. Serial order: A parallel, distributed processing approach. In Advances in Connectionist Theory: Speech. Lawrence Erlbaum Associates. find
  12. Nal Kalchbrenner, Edward Grefenstette, and Phil Blunsom. 2014. A convolutional neural network for modelling sentences. In ACL, pages 655–665, Baltimore, Maryland, June. ACL.
  13. John F. Kelley. 1984. An iterative design methodology for user-friendly natural language office information applications. ACM Transaction on Information Systems. find
  14. Yoon Kim. 2014. Convolutional neural networks for sentence classification. In EMNLP, pages 1746– 1751, Doha, Qatar, October. ACL.
  15. Jiwei Li, Michel Galley, Chris Brockett, Jianfeng Gao, and Bill Dolan. 2016. A diversity-promoting objective function for neural conversation models. In NAACL-HLT, pages 110–119, San Diego, California, June. ACL.
  16. Wang Ling, Phil Blunsom, Edward Grefenstette, Karl Moritz Hermann, Tomáš Kočiský, Fumin Wang, and Andrew Senior. 2016. Latent predictor networks for code generation. In ACL, pages 599–609, Berlin, Germany, August. ACL. find
  17. Tomáš Mikolov, Martin Karafiat, Lukáš Burget, Jan Černocký, and Sanjeev Khudanpur. 2010. Recurrent neural network based language model. In Interspeech, pages 1045–1048, Makuhari, Japan. ISCA. find
  18. Nikola Mrkšić, Diarmuid Ó Séaghdha, Blaise Thomson, Milica Gašić, Pei-Hao Su, David Vandyke, Tsung-Hsien Wen, and Steve Young. 2015. Multi-domain dialog state tracking using recurrent neural networks. In ACL, pages 794–799, Beijing, China, July. ACL. find
  19. Nikola Mrkšić, Diarmuid Ó Séaghdha, Tsung-Hsien Wen, Blaise Thomson, and Steve Young. 2016. Neural belief tracker: Data-driven dialogue state tracking. arXiv preprint:1606.03777. find
  20. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. Bleu: A method for automatic evaluation of machine translation. In ACL, pages 311–318, Stroudsburg, PA, USA. ACL.
  21. Iulian Vlad Serban, Ryan Lowe, Laurent Charlin, and Joelle Pineau. 2015a. A survey of available corpora for building data-driven dialogue systems. arXiv preprint:1512.05742. find
  22. Iulian Vlad Serban, Alessandro Sordoni, Yoshua Bengio, Aaron C. Courville, and Joelle Pineau. 2015b. Hierarchical neural network generative models for movie dialogues. arXiv preprint:1507.04808.
  23. Lifeng Shang, Zhengdong Lu, and Hang Li. 2015. Neural responding machine for short-text conversation. In ACL, pages 1577–1586, Beijing, China, July. ACL.
  24. Pei-Hao Su, David Vandyke, Milica Gasic, Dongho Kim, Nikola Mrksic, Tsung-Hsien Wen, and Steve J. Young. 2015. Learning from real users: rating dialogue success with neural networks for reinforcement learning in spoken dialogue systems. In Interspeech, pages 2007–2011, Dresden, Germany. ISCA. find
  25. Pei-Hao Su, Milica Gasic, Nikola Mrkšić, Lina M. Rojas Barahona, Stefan Ultes, David Vandyke, Tsung-Hsien Wen, and Steve Young. 2016. On-line active reward learning for policy optimisation in spoken dialogue systems. In ACL, pages 2431–2441, Berlin, Germany, August. ACL. find
  26. Steve Young, Milica Gašić, Simon Keizer, François Mairesse, Jost Schatzmann, Blaise Thomson, and Kai Yu. 2010. The hidden information state model: A practical framework for pomdp-based spoken dialogue management. Computer, Speech and Language. find
  27. Sainbayar Sukhbaatar, arthur szlam, Jason Weston, and Rob Fergus. 2015. End-to-end memory networks. In NIPS, pages 2440–2448. Curran Associates, Inc., Montreal, Canada. find
  28. Steve Young, Milica Gašić, Blaise Thomson, and Jason D. Williams. 2013. Pomdp-based statistical spoken dialog systems: A review. Proceedings of the IEEE. find
  29. Ilya Sutskever, Oriol Vinyals, and Quoc V. Le. 2014. Sequence to sequence learning with neural networks. In NIPS, pages 3104–3112, Montreal, Canada. MIT Press.
  30. David R. Traum, 1999. Foundations of Rational Agency, chapter Speech Acts for Dialogue Agents. Springer. find
  31. Oriol Vinyals and Quoc V. Le. 2015. A neural conversational model. In ICML Deep Learning Workshop, Lille, France.
  32. Oriol Vinyals, Meire Fortunato, and Navdeep Jaitly. 2015. Pointer networks. In NIPS, pages 2692–2700, Montreal, Canada. Curran Associates, Inc. find
  33. Tsung-Hsien Wen, Aaron Heidel, Hung yi Lee, Yu Tsao, and Lin-Shan Lee. 2013. Recurrent neural network based language model personalization by social network crowdsourcing. In Interspeech, pages 2007– 2011, Lyon France. ISCA. find
  34. Tsung-Hsien Wen, Milica Gašić, Dongho Kim, Nikola Mrkšić, Pei-Hao Su, David Vandyke, and Steve Young. 2015a. Stochastic language generation in dialogue using recurrent neural networks with convolutional sentence reranking. In SIGdial, pages 275–284, Prague, Czech Republic, September. ACL.
  35. Tsung-Hsien Wen, Milica Gašić, Nikola Mrkšić, Pei-Hao Su, David Vandyke, and Steve Young. 2015b. Semantically conditioned lstm-based natural language generation for spoken dialogue systems. In EMNLP, pages 1711–1721, Lisbon, Portugal, September. ACL.
  36. Tsung-Hsien Wen, Milica Gašić, Nikola Mrkšić, Pei-Hao Su, David Vandyke, and Steve Young. 2016. Multi-domain neural network language generation for spoken dialogue systems. In NAACL-HLT, pages 120–129, San Diego, California, June. ACL.
  37. Kaisheng Yao, Baolin Peng, Yu Zhang, Dong Yu, Geoffrey Zweig, and Yangyang Shi. 2014. Spoken language understanding using long short-term memory neural networks. In IEEE SLT, pages 189–194, December. find
  38. Pengcheng Yin, Zhengdong Lu, Hang Li, and Ben Kao. 2015. Neural enquirer: Learning to query tables. arXiv preprint:1512.00965. — [arXiv:1512.00965](https://arxiv.org/abs/1512.00965) find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/wen-2017-network-based-end-end-trainable/