Reinforcement Learning for Bandit Neural Machine Translation with Simulated Human Feedback
Open accessThe DOI matches its Crossref record; 43 of the 43 references it cites were found in a registry or an index.
Abstract
Machine translation is a natural candidate problem for reinforcement learning from human feedback: users provide quick, dirty ratings on candidate translations to guide a system to improve.Yet, current neural machine translation training focuses on expensive human-generated reference translations.We describe a reinforcement learning algorithm that improves neural machine translation systems from simulated human feedback.Our algorithm combines the advantage actor-critic algorithm (Mnih et al., 2016) with the attention-based neural encoderdecoder architecture (Luong et al., 2015).This algorithm (a) is well-designed for problems with a large action space and delayed rewards, (b) effectively optimizes traditional corpus-level machine translation metrics, and (c) is robust to skewed, high-variance, granular feedback modeled after actual human behaviors.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
Effective Approaches to Attention-based Neural Machine Translation
Luong, T., Pham, H., & Manning, C. D. (2015) · Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d15-1166 · Cited by 8,382
Asynchronous Methods for Deep Reinforcement Learning
Mnih, V., Badia, A. P., Mirza, M., Graves, A., Lillicrap, T. P., Harley, T., Silver, D., & Kavukcuoglu, K. (2016) · arXiv · doi:10.48550/arxiv.1602.01783 · Cited by 7,132
Moses: open source toolkit for statistical machine translation
Koehn, P., Zens, R., Dyer, C., Bojar, O., Constantin, A., Herbst, E., Hoang, H., Birch, A., Callison-Burch, C., Federico, M., et al. (2007) · Proceedings of the 45th Annual Meeting of the ACL on Interactive Poster and Demonstration Sessions - ACL '07 · doi:10.3115/1557769.1557821 · Cited by 4,735 · cited 1×
“For English and German, we tokenize and clean sentences using Moses (Koehn et al., 2007).”
MethodNeural Architecture Search with Reinforcement Learning
Zoph, B., & Le, Q. V. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1611.01578 · Cited by 3,974
Preston, C. C., & Colman, A. M. (2000) · Acta Psychologica · doi:10.1016/s0001-6918(99)00050-5 · Cited by 2,106
Cheap and fast---but is it good?: evaluating non-expert annotations for natural language tasks
Snow, R., O'Connor, B., Jurafsky, D., & Ng, A. Y. (2008) · Proceedings of the Conference on Empirical Methods in Natural Language Processing - EMNLP '08 · doi:10.3115/1613715.1613751 · Cited by 1,887
Explaining collaborative filtering recommendations
Herlocker, J. L., Konstan, J. A., & Riedl, J. (2000) · Proceedings of the 2000 ACM conference on Computer supported cooperative work · doi:10.1145/358916.358995 · Cited by 1,731
Sequence Level Training with Recurrent Neural Networks
Ranzato, M., Chopra, S., Auli, M., & Zaremba, W. (2015) · arXiv · doi:10.48550/arxiv.1511.06732 · Cited by 1,426
Deep Reinforcement Learning for Dialogue Generation
Li, J., Monroe, W., Ritter, A., Jurafsky, D., Galley, M., & Gao, J. (2016) · Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d16-1127 · Cited by 1,204
Simple statistical gradient-following algorithms for connectionist reinforcement learning
Williams, R. J. (1992) · Machine Learning · doi:10.1023/a:1022672621406 · Cited by 817
Sequence Level Training with Recurrent Neural Networks
Ranzato, M., Chopra, S., Auli, M., & Zaremba, W. (2016) · International Conference on Learning Representations · Cited by 572
WIT3: Web Inventory of Transcribed and Translated Talks
Cettolo, M., Girardi, C., & Federico, M. (2012) · Cited by 543
Estimation with Heteroscedastic Error Terms
Park, R. E. (1966) · Econometrica · doi:10.2307/1910108 · Cited by 426
An Actor-Critic Algorithm for Sequence Prediction
Bahdanau, D., Brakel, P., Xu, K., Goyal, A., Lowe, R., Pineau, J., Courville, A., & Bengio, Y. (2016) · arXiv · doi:10.48550/arxiv.1607.07086 · Cited by 423
Teachable robots: Understanding human teaching behavior to build more effective robot learners
Thomaz, A. L., & Breazeal, C. (2008) · Artificial Intelligence · doi:10.1016/j.artint.2007.09.009 · Cited by 405
Deep Reinforcement Learning for Dialogue GenerationPreprint
Li, J., Monroe, W., Ritter, A., et al. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1606.01541 · Cited by 405
Most likely heteroscedastic Gaussian process regression
Kersting, K., Plagemann, C., Pfaff, P., & Burgard, W. (2007) · Proceedings of the 24th international conference on Machine learning · doi:10.1145/1273496.1273546 · Cited by 311
Device Placement Optimization with Reinforcement Learning
Mirhoseini, A., Pham, H., Le, Q. V., Steiner, B., Larsen, R., Zhou, Y., Kumar, N., Norouzi, M., Bengio, S., & Dean, J. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1706.04972 · Cited by 299
Optimizing Chinese word segmentation for machine translation performance
Chang, P.-C., Galley, M., & Manning, C. D. (2008) · Proceedings of the Third Workshop on Statistical Machine Translation - StatMT '08 · doi:10.3115/1626394.1626430 · Cited by 296
Thomaz, A. L., & Breazeal, C. (2006) · Cited by 259
Efficient bandit algorithms for online multiclass prediction
Kakade, S. M., Shalev-Shwartz, S., & Tewari, A. (2008) · Proceedings of the 25th international conference on Machine learning - ICML '08 · doi:10.1145/1390156.1390212 · Cited by 168
Impact of data characteristics on recommender systems performance
Adomavicius, G., & Zhang, J. (2012) · ACM Transactions on Management Information Systems · doi:10.1145/2151163.2151166 · Cited by 164
Report on the 11th IWSLT Evaluation Campaign, IWSLT 2014
Cettolo, M., Niehues, J., Stüker, S., Bentivogli, L., & Federico, M. (2014) · Cited by 158
Online human training of a myoelectric prosthesis controller via actor-critic reinforcement learning
Pilarski, P. M., Dawson, M. R., Degris, T., Fahimi, F., Carey, J. P., & Sutton, R. S. (2011) · 2011 IEEE International Conference on Rehabilitation Robotics · doi:10.1109/icorr.2011.5975338 · Cited by 153
Can machine translation systems be evaluated by the crowd alone
GRAHAM, Y., BALDWIN, T., MOFFAT, A., & ZOBEL, J. (2017) · Natural Language Engineering · doi:10.1017/s1351324915000339 · Cited by 140
A social reinforcement learning agent
Isbell, C., Shelton, C. R., Kearns, M., Singh, S., & Stone, P. (2001) · Proceedings of the fifth international conference on Autonomous agents · doi:10.1145/375735.376334 · Cited by 138
Dynamic Reward Shaping: Training a Robot by Voice
Tenorio-González, A. C., Morales, E. F., & Villaseñor-Pineda, L. (2010) · Lecture notes in computer science · doi:10.1007/978-3-642-16952-6_49 · Cited by 106
Don’t Until the Final Verb Wait: Reinforcement Learning for Simultaneous Machine Translation
Grissom II, A., He, H., Boyd-Graber, J., Morgan, J., & Daumé III, H. (2014) · Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.3115/v1/d14-1140 · Cited by 98
Self-citationThe IWSLT 2015 Evaluation Campaign
Cettolo, M., Niehues, J., Stüker, S., Bentivogli, L., Cattoni, R., & Federico, M. (2015) · Repository KITopen (Karlsruhe Institute of Technology) · doi:10.5281/zenodo.1326521 · Cited by 95
Personalized Machine Translation: Preserving Original Author Traits
Rabinovich, E., Patel, R. N., Mirkin, S., Specia, L., & Wintner, S. (2017) · Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers · doi:10.18653/v1/e17-1101 · Cited by 91
Learning to Search Better Than Your Teacher
Chang, K.-W., Krishnamurthy, A., Agarwal, A., Daumé, H., & Langford, J. C. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1502.02206 · Cited by 84
Motivating Personality-aware Machine Translation
Mirkin, S., Nowson, S., Brun, C., & Pérez, J. (2015) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/d15-1130 · Cited by 72
Active learning in heteroscedastic noise
Antos, A., Grover, V., & Szepesvári, C. (2010) · Theoretical Computer Science · doi:10.1016/j.tcs.2010.04.007 · Cited by 71
Interpretese vs. Translationese: The Uniqueness of Human Strategies in Simultaneous Interpretation
He, H., Boyd-Graber, J., & Daumé III, H. (2016) · Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies · doi:10.18653/v1/n16-1111 · Cited by 59
Self-citationA Strategy-Aware Technique for Learning Behaviors from Discrete Human Feedback
Loftin, R., MacGlashan, J., Peng, B., Taylor, M. E., Littman, M. L., Huang, J. J. S., & Roberts, D. L. (2014) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v28i1.8839 · Cited by 55
Bandit Structured Prediction for Neural Sequence-to-Sequence Learning
Kreutzer, J., Sokolov, A., & Riezler, S. (2017) · Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p17-1138 · Cited by 35
Learning Structured Predictors from Bandit Feedback for Interactive NLP
Sokolov, A., Kreutzer, J., Lo, C., & Riezler, S. (2016) · Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p16-1152 · Cited by 28
Stochastic Structured Prediction under Bandit FeedbackPreprint
Sokolov, A., Kreutzer, J., Lo, C., & Riezler, S. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1606.00739 · Cited by 25
Syntax-based Rewriting for Simultaneous Machine Translation
He, H., Grissom, A., Morgan, J., Boyd-Graber, J. L., & Daumé, H. (2015) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/d15-1006 · Cited by 24
Self-citationBandit Structured Prediction for Learning from Partial Feedback in Statistical Machine Translation
Sokolov, A., Riezler, S., & Urvoy, T. (2016) · arXiv · doi:10.48550/arxiv.1601.04468 · Cited by 14
Crowdsourced Monolingual Translation
Hu, C., Resnik, P., & Bederson, B. B. (2014) · ACM Transactions on Computer-Human Interaction · doi:10.1145/2627751 · Cited by 8
A Coactive Learning View of Online Structured Prediction in Statistical Machine Translation
Sokolov, A., Riezler, S., & Cohen, S. B. (2015) · Conference on Computational Natural Language Learning · doi:10.18653/v1/k15-1001 · Cited by 4
Research works printed but not linked: 6
- Boxing Chen and Colin Cherry. 2014. A systematic comparison of smoothing techniques for sentencelevel bleu. In Association for Computational Linguistics (ACL) .
- John Langford and Tong Zhang. 2008. The epochgreedy algorithm for multi-armed bandits with side information. In Advances in Neural Information Processing Systems (NIPS) .
- Rensis Likert. 1932. A technique for the measurement of attitudes. Archives of Psychology 22(140):1-55.
- Minh-Thang Luong, Hieu Pham, and Christopher D Manning. 2015. Effective approaches to attentionbased neural machine translation. In Empirical Methods in Natural Language Processing (EMNLP) .
- Rebecca J Passonneau and Bob Carpenter. 2014. The benefits of a model of annotation. Transactions of the Association for Computational Linguistics (TACL) 2:311-326.
- Ronald J Williams. 1992. Simple statistical gradientfollowing algorithms for connectionist reinforcement learning. Machine learning 8(3-4):229-256.
Cited by
Learning to summarize from human feedback
Stiennon, N., Ouyang, L., Wu, J., Ziegler, D. M., Lowe, R., Voss, C., Radford, A., Amodei, D., & Christiano, P. (2020) · arXiv · doi:10.48550/arxiv.2009.01325 · Cited by 550
“There is also a rich literature on using RL to optimize automatic metrics for NLP tasks, such as ROUGE for summarization [50, 65, 45, 15, 19], BLEU for translation [50, 66, 1, 43], and other domains [61, 27, 26].”
BackgroundFine-Tuning Language Models from Human Preferences
Ziegler, D. M., Stiennon, N., Wu, J., Brown, T. B., Radford, A., Amodei, D., Christiano, P., & Irving, G. (2019) · arXiv · doi:10.48550/arxiv.1909.08593 · Cited by 413
Yang, Q., Zhang, Y., Dai, W., & Pan, S. J. (2020) · Cambridge University Press eBooks · doi:10.1017/9781139061773 · Cited by 254
A Study of Reinforcement Learning for Neural Machine Translation
Wu, L., Tian, F., Qin, T., Lai, J., & Liu, T.-Y. (2018) · Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/d18-1397 · Cited by 157
Reinforcement-Learning-Guided Source Code Summarization Using Hierarchical Attention
Wang, W., Zhang, Y., Sui, Y., Wan, Y., Zhao, Z., Wu, J., Yu, P. S., & Xu, G. (2022) · IEEE Transactions on Software Engineering · doi:10.1109/tse.2020.2979701 · Cited by 103
Metadata
- Authors
- Nguyen, K., Daumé III, H., & Boyd-Graber, J.
- Affiliations
- University of Maryland, College Park (US)
- Microsoft Research New York City (United States) (US)
- Venue
- Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing
- Published
- 2017
- Pages
- 1464-1474
- OpenAlex
- W2737114849
- Cited by
- 110
- References
- 40
- Citations by year
- 2026: 2 · 2025: 4 · 2024: 6 · 2023: 12 · 2022: 5 · 2021: 13 · 2020: 36 · 2019: 16 · 2018: 14 · 2017: 2
- FWCI
- 8.30
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Conference on Empirical Methods in Natural Language Processing (EMNLP) (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- National Science Foundation (1422492, 1618193, 1320538, 1564275, 1409287, 1652666)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Reinforcement Learning in Robotics
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Keywords
- reinforcement learning from human feedback, neural machine translation, Advantage Actor-Critic, attention-based encoder-decoder, delayed rewards, large action spaces
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.