Abstract
Deep learning has recently been adopted for the task of natural language generation (NLG) and shown remarkable results. However, learning can go awry when the input dataset is too small or not well balanced with regards to the examples it contains for various input sequences. This is relevant to naturally occurring datasets such as many that were not prepared for the task of natural language processing but scraped off the web and originally prepared for a different purpose. As a mitigation to the problem of unbalanced training data, we therefore propose to decompose a large natural language dataset into several subsets that “talk about” the same thing. We show that the decomposition helps to focus each learner’s attention during training. Results from a proof-of-concept study show 73% times faster learning over a flat model and better results.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
References
Angeli, G., Liang, P., Klein, D.: A simple domain-independent probabilistic approach to generation. In: Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP), Cambridge, Massachusetts (2010)
Belz, A., Gatt, A.: Intrinsic vs. extrinsic evaluation measures for referring expression generation. In: Proceedings of the 46th Annual Meeting of the Association for Computational Linguistics (ACL), Columbus, OH, USA (2008)
Bengio, Y., Simard, P., Frasconi, P.: Learning long-term dependencies with gradient descent is difficult. IEEE Trans. Neural Netw. 5(2), 157–166 (1994)
Cho, K., van Merrienboer, B., Gülçehre, Ç., Bougares, F., Schwenk, H., Bengio, Y.: Learning phrase representations using RNN encoder-decoder for statistical machine translation. In: Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), Doha, Qatar (2014)
Cuayáhuitl, H., Dethlefs, N., Hastie, H., Liu, X.: Training a statistical surface realiser from automatic slot labelling. In: Proceedings of the IEEE Workshop on Spoken Language Technology (SLT), South Lake Tahoe, USA (2014)
Dethlefs, N., Cuayáhuitl, H.: Hierarchical reinforcement learning and hidden markov models for task-oriented natural language generation. In: Proceedings of the 49th Annual Conference of the Association for Computational Linguistics (ACL-HLT), Short Papers, Portland, OR, USA (2011)
Dethlefs, N., Cuayáhuitl, H.: Hierarchical reinforcement learning for situated natural language generation. Natl. Lang. Eng. 21, 391–435 (2015)
Dethlefs, N., Hastie, H., Cuayáhuitl, H., Lemon, O.: Conditional random fields for responsive surface realisation. In: Proceedings of the 51st Annual Meeting of the Association for Computational Linguistics (ACL), Sofia, Bulgaria (2013)
Dusek, O., Jurcicek, F.: Sequence-to-sequence generation for spoken dialogue via deep syntax trees and strings. In: Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL), Berlin, Germany (2016)
Graves, A.: Generating sequences with recurrent neural networks. CoRR abs/1308.0850 (2013). http://arxiv.org/abs/1308.0850
Konstas, I., Lapata, M.: Unsupervised concept-to-text generation with hypergraphs. In: Proceedings of the North American Chapter of the Association for Computational Linguistics (NAACL), Montreal, Canada (2012)
Liang, P., Jordan, M., Klein, D.: Learning semantic correspondences with less supervision. In: Proceedings of the 47th Annual Meeting of the Association for Computational Linguistics (ACL), Singapore (2009)
Mairesse, F., Jurčíček, F., Keizer, S., Thomson, B., Yu, K., Young, S.: Phrase-based statistical language generation using graphical models and active learning. In: Proceedings of the 48th Annual Meeting of the Association of Computational Linguistics (ACL), Uppsala, Sweden (2010)
Mei, H., Bansal, M., Walker, M.: What to talk about and how? Selective Generation using LSTMs with coarse-to-fine alignment. In: Proceedings of the North American Chapter of the Association for Computational Linguistics (NAACL), San Diego, CA, USA (2016)
Morin, F., Bengio, Y.: Hierarchical probabilistic neural network language model. In: Proceedings of the International Conference on Artificial Intelligence and Statistics (AISTATS), pp. 246–252 (2005)
Novikova, J., Rieser, V.: The aNALoGuE Challenge: Non Aligned Language GEneration. In: Proceedings of the 9th International Natural Language Generation Conference (INLG) (2016)
Papineni, K., Roukos, S., Ward, T., Zhu, W.J.: BLEU: a method for automatic evaluation of machine translation. In: Proceedings of the 40th Annual Meeting on Association for Computational Linguistics (ACL), Association for Computational Linguistics, pp. 311–318 (2001)
Reiter, E., Dale, R.: Building Natural Language Generation Systems. Cambridge University Press, New York (2000)
Snyder, B., Barzilay, R.: Database-text alignment via structured multilabel classification. In: Proceedings of 20th International Joint Conference on Artificial Intelligence (IJCAI), Hyderabad, India (2007)
Sutskever, I., Vinyals, O., Le, Q.V.: Sequence to sequence learning with neural networks. Adv. Neural Inf. Process. Syst. (NIPS) 27, 3104–3112 (2014)
Turner, A.P., Caves, L.S., Stepney, S., Tyrrell, A.M., Lones, M.A.: Artificial epigenetic networks: automatic decomposition of dynamical control tasks using topological self-modification. IEEE Trans. Neural Netw. Learn. Syst. (2016)
Turner, A.P., Lones, M.A., Fuente, L.A., Stepney, S., Caves, L.S., Tyrrell, A.M.: The artificial epigenetic network. In: 2013 IEEE International Conference on Evolvable Systems (ICES), pp. 66–72. IEEE (2013)
Walker, M., Stent, A., Mairesse, F., Prasad, R.: Individual and domain adaptation in sentence planning for dialogue. J. Artif. Intell. Res. 30(1), 413–456 (2007)
Wen, T.H., Gašić, M., Mrkšić, N., Su, P.H., Vandyke, D., Young, S.: Semantically conditioned LSTM-based natural language generation for spoken dialogue systems. In: Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing (EMNLP) (2015)
Acknowledgements
We acknowledge the VIPER high-performance computing facility of the University of Hull and its support team. We are also grateful for Nvidia’s donation of a Titan X Pascal graphics card for our work on deep learning.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Dethlefs, N., Turner, A. (2017). Deep Text Generation – Using Hierarchical Decomposition to Mitigate the Effect of Rare Data Points. In: Gracia, J., Bond, F., McCrae, J., Buitelaar, P., Chiarcos, C., Hellmann, S. (eds) Language, Data, and Knowledge. LDK 2017. Lecture Notes in Computer Science(), vol 10318. Springer, Cham. https://doi.org/10.1007/978-3-319-59888-8_25
Download citation
DOI: https://doi.org/10.1007/978-3-319-59888-8_25
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-59887-1
Online ISBN: 978-3-319-59888-8
eBook Packages: Computer ScienceComputer Science (R0)