Non-linear Learning for Statistical Machine Translation

被引:0
|
作者
Huang, Shujian [1 ]
Chen, Huadong [1 ]
Dai, Xinyu [1 ]
Chen, Jiajun [1 ]
机构
[1] Nanjing Univ, State Key Lab Novel Software Technol, Nanjing 210023, Jiangsu, Peoples R China
基金
中国国家自然科学基金;
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Modern statistical machine translation (SMT) systems usually use a linear combination of features to model the quality of each translation hypothesis. The linear combination assumes that all the features are in a linear relationship and constrains that each feature interacts with the rest features in an linear manner, which might limit the expressive power of the model and lead to a under-fit model on the current data. In this paper, we propose a non-linear modeling for the quality of translation hypotheses based on neural networks, which allows more complex interaction between features. A learning framework is presented for training the non-linear models. We also discuss possible heuristics in designing the network structure which may improve the non-linear learning performance. Experimental results show that with the basic features of a hierarchical phrase-based machine translation system, our method produce translations that are better than a linear model.
引用
收藏
页码:825 / 835
页数:11
相关论文
共 50 条
  • [1] Editorial: Non-linear analysis and machine learning in cardiology
    Dierckx, Hans
    Zhao, Xiaopeng
    Tolkacheva, Elena G.
    FRONTIERS IN PHYSIOLOGY, 2023, 14
  • [2] Non-linear development in statistical learning of visual orthographic regularities
    Duan, Rujun
    Sun, Qi
    Tong, Xiuhong
    NPJ SCIENCE OF LEARNING, 2025, 10 (01)
  • [3] Online Learning for Statistical Machine Translation
    Ortiz-Martinez, Daniel
    COMPUTATIONAL LINGUISTICS, 2016, 42 (01) : 121 - 161
  • [4] Machine Learning Guided Discovery of Non-Linear Optical Materials
    Mondal, Sownyak
    Hammad, Raheel
    ADVANCED THEORY AND SIMULATIONS, 2025, 8 (02)
  • [5] Taming Non-Linear Dynamics and Turbulence with Machine Learning Control
    Maceda, Guy Y. Cornejo
    Lusseyran, Francois
    Noack, Bernd R.
    ERCIM NEWS, 2020, (122): : 32 - 33
  • [6] A Bayesian non-linear method for feature selection in machine translation quality estimation
    Shah, Kashif
    Cohn, Trevor
    Specia, Lucia
    MACHINE TRANSLATION, 2015, 29 (02) : 101 - 125
  • [7] Learning spatiotemporal statistical shape models for non-linear dynamic anatomies
    Adams, Jadie
    Khan, Nawazish
    Morris, Alan
    Elhabian, Shireen
    FRONTIERS IN BIOENGINEERING AND BIOTECHNOLOGY, 2023, 11
  • [8] THE STATISTICAL THEORY OF A NON-LINEAR FIELD
    TERLETSKII, JP
    DOKLADY AKADEMII NAUK SSSR, 1960, 133 (03): : 568 - 571
  • [9] STATISTICAL PROPERTIES OF A NON-LINEAR STRING
    IZRAILEV, FM
    CHIRIKOV, BV
    DOKLADY AKADEMII NAUK SSSR, 1966, 166 (01): : 57 - &
  • [10] On the Relationship of Emergence and Non-Linear Dynamics to Machine Learning and Synthetic Consciousness
    Easttom, Chuck
    PROCEEDINGS OF THE EUROPEAN CONFERENCE ON THE IMPACT OF ARTIFICIAL INTELLIGENCE AND ROBOTICS (ECIAIR 2019), 2019, : 122 - 130