Abstract
Replacing the traditional cross-entropy loss with BLEU as the optimization objective is a successful application of reinforcement learning (RL) in neural machine translation (NMT). However, a considerable weakness of the approach is that the monotonic optimization of BLEU’s training algorithm ignores the semantic fluency of the translation. One phenomenon is an incomprehensible translation accompanied by an ideal BLEU. In addition, sampling inefficiency as a common shortcoming of RL is more prominent in NMT. In this study, we address these issues in two ways. (1) We use the annealing schedule algorithm to add semantic evaluation for reinforcement training as part of the training objective. (2) We further attach a value iteration network to RL to transform the reward into a decision value, thereby making model training highly targeted and efficient. We use our approach on three representative language machine translation tasks, including low resource Mongolian-Chinese, agglutinative Japanese-English, and common task English-Chinese. Experiments show that our approach achieves significant improvements over the strong baselines, besides, it also saves nearly one-third of training time on different tasks.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
References
Bahdanau, D., Cho, K., Bengio, Y.: Neural machine translation by jointly learning to align and translate. In: International Conference on Learning Representations (ICLR), pp. 473–484 (2014)
Gehring, J., Auli, M., Grangier, D., et al.: Convolutional sequence to sequence learning. In: International Conference on Machine Learning (ICML), pp. 1243–1252 (2017)
Hessel, M., Modayil, J., Van Hasselt, H., et al.: Rainbow: combining improvements in deep reinforcement learning. In: National Conference on Artificial Intelligence (NCAI), pp. 3215–3222 (2018)
Hui, K.P., Bean, N., Kraetzl, M., et al.: The cross-entropy method for network reliability estimation. Ann. Oper. Res. 134(1), 101 (2005)
Kalchbrenner, N., Blunsom, P.: Recurrent continuous translation models. In: Conference on Empirical Methods in Natural Language Processing (EMNLP), pp. 1700–1709 (2013)
Kunchukuttan, A., Bhattacharyya, P.: Learning variable length units for SMT between related languages via Byte Pair Encoding. In: Conference on Empirical Methods in Natural Language Processing (EMNLP), pp. 14–24 (2017)
Mikolov, T., Kombrink, S., Burget, L., et al.: Extensions of recurrent neural network language model. In: IEEE International Conference on Acoustics, Speech, and Signal Processing, pp. 5528–5531 (2011)
Morin, F., Bengio, Y.: Hierarchical probabilistic neural network language model. In: International Conference on Artificial Intelligence and Statistics (AISTATS), pp. 246–252 (2005)
Nogueira, R., Cho, K.: WebNav: a new large-scale task for natural language based sequential decision making. In: Conference and Workshop on Neural Information Processing Systems (NIPS), pp. 177–186 (2016)
Ranzato, M., Chopra, S., Auli, M., Zaremba, W.: Sequence level training with recurrent neural networks (2015). arXiv:1511.06732
Rush, A.M., Chopra, S., Weston, J.: A neural attention model for abstractive sentence summarization. In: Conference on Empirical Methods in Natural Language Processing (EMNLP), pp. 379–389 (2015)
Schwenker, F., Palm, G.: Artificial development by reinforcement learning can benefit from multiple motivations. Front. Robot. AI 6(6) (2019)
Sennrich, R., Haddow, B., Birch, A.: Neural machine translation of rare words with subword units. In: Association for Computational Linguistics (ACL), pp. 1715–1725 (2015)
Shen, S., Cheng, Y., He, Z., et al.: Minimum risk training for neural machine translation. In: Association for Computational Linguistics (ACL), pp. 1683–1692 (2016)
Smith, D.K.: Dynamic programming and optimal control. J. Oper. Res. Soc. 47(06), 833–834 (1996)
Sunmola, F.T., Wyatt, J.L.: Model transfer for Markov decision tasks via parameter matching. In: Proceedings of the 25th Workshop of the UK Planning and Scheduling Special Interest Group (PlanSIG 2006), pp. 77–86 (2006)
Sutskever, I., et al.: Sequence to sequence learning with neural networks. In: Conference and Workshop on Neural Information Processing Systems (NIPS), pp. 3104–3112 (2014)
Vaswani, A., Shazeer, N., Parmar, N., et al.: Attention is all you need. In: Conference and Workshop on Neural Information Processing Systems (NIPS), pp. 5998–6008 (2017)
Volodymyr, M., Koray, K., et al.: Human-level control through deep reinforcement learning. Nature 518(7540), 529 (2015)
From Wikipedia Dynamic Programming and Markov Processes: Markov decision process. J. Oper. Res. Soc. 112(4), 217–243 (2010)
Wu, L., Tian, F., Qin, T., et al.: A study of reinforcement learning for neural machine translation. In: Conference on Empirical Methods in Natural Language Processing (EMNLP), pp. 3215–3222 (2018)
Zaremba, W., Sutskever, I.: Reinforcement learning neural turing machines-revised (2015). arXiv:1505.00521
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Ji, Y., Hou, H., Chen, J., Wu, N. (2019). Training with Additional Semantic Constraints for Enhancing Neural Machine Translation. In: Nayak, A., Sharma, A. (eds) PRICAI 2019: Trends in Artificial Intelligence. PRICAI 2019. Lecture Notes in Computer Science(), vol 11670. Springer, Cham. https://doi.org/10.1007/978-3-030-29908-8_24
Download citation
DOI: https://doi.org/10.1007/978-3-030-29908-8_24
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-29907-1
Online ISBN: 978-3-030-29908-8
eBook Packages: Computer ScienceComputer Science (R0)