A Shared Task on Bandit Learning for Machine Translation
作者： Stefan Riezler , Hagen Fürstenau , Julia Kreutzer , Kellen Sunderland , Witold Szymaniak
DOI:
关键词: Natural language processing 、 Computer science 、 Translation (geometry) 、 Quality (business) 、 Machine translation 、 Sequence 、 Task (project management) 、 Artificial intelligence 、 Multi-task learning
摘要: We introduce and describe the results of a novel shared task on bandit learning for machine translation. The was organized jointly by Amazon Heidelberg University first time at Second Conference Machine Translation (WMT 2017). goal is to encourage research translation from weak user feedback instead human references or post-edits. On each sequence rounds, system required propose an input, receives real-valued estimate quality proposed learning. This paper describes task's evaluation setup, using services hosted Web Services (AWS), data metrics, various architectures protocols.
core.ac.uk
arxiv.org参考文章(27)
Matthew D. Zeiler, ADADELTA: An Adaptive Learning Rate Method arXiv: Learning. ,(2012)
Diederik P. Kingma, Jimmy Ba, Adam: A Method for Stochastic Optimization arXiv: Learning. ,(2014)
Jianfeng Gao, Xiaodong He, Amittai Axelrod, Domain Adaptation via Pseudo In-Domain Data Selection empirical methods in natural language processing. pp. 355- 362 ,(2011)
Olivier Chapelle, Eren Manavoglu, Romer Rosales, Simple and Scalable Response Prediction for Display Advertising ACM Transactions on Intelligent Systems and Technology. ,vol. 5, pp. 61- ,(2014) , 10.1145/2532128
H. Brendan McMahan, Adam Tauman Kalai, Abraham D. Flaxman, Online convex optimization in the bandit setting: gradient descent without a gradient symposium on discrete algorithms. pp. 385- 394 ,(2005) , 10.5555/1070432.1070486
Csaba Szepesvari, Algorithms for Reinforcement Learning ,(2010)
Ilya Sutskever, Geoffrey Hinton, Alex Krizhevsky, Ruslan Salakhutdinov, Nitish Srivastava, Dropout: a simple way to prevent neural networks from overfitting Journal of Machine Learning Research. ,vol. 15, pp. 1929- 1958 ,(2014)
Zhifei Li, Jason Eisner, First- and Second-Order Expectation Semirings with Applications to Minimum-Risk Training on Translation Forests empirical methods in natural language processing. pp. 40- 51 ,(2009) , 10.3115/1699510.1699517
Chin-Yew Lin, Franz Josef Och, Automatic Evaluation of Machine Translation Quality Using Longest Common Subsequence and Skip-Bigram Statistics meeting of the association for computational linguistics. pp. 605- 612 ,(2004) , 10.3115/1218955.1219032
Thang Luong, Ilya Sutskever, Quoc Le, Oriol Vinyals, Wojciech Zaremba, Addressing the Rare Word Problem in Neural Machine Translation Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). ,vol. 1, pp. 11- 19 ,(2015) , 10.3115/V1/P15-1002