%0 Journal Article
%T 面向序列迁移学习的似然比模型选择方法<br>Model selection with likelihood ratio for sequence transfer learning
%A 孙世昶
%A 林鸿飞
%A 孟佳娜
%A 刘洪波&lt
%A br&gt
%A SUN Shi-chang
%A LIN Hong-fei
%A MENG Jia-na
%A LIU Hong-bo
%J 山东大学学报（理学版）
%D 2017
%R 10.6040/j.issn.1671-9352.5.2016.023
%X 摘要： 为了解决迁移学习的欠适配问题,将粒模型作为候选模型的集合,通过模型选择的方式引入目标域的辅助模型中包含的标注规则,提出粒模型推断中基于似然比的模型选择方法(likelihood ratio model selection, LRMS),实现了辅助模型与粒模型的融合。LRMS保持基于Viterbi算法的标注模型对整条序列进行计算的模式,避免了候选标注器对上下文关系的破坏。通过大量词性标注实验表明LRMS在每个迁移学习任务中都有准确率的提高,从而证明似然比模型选择是一种有效的解决欠适配问题的方法。<br>Abstract: To solve the under-adaptation problem of transfer learning,in this paper the granular model is used as a set of candidate models, and labeling rules contained in minor for target domain models is introduced by a model selection method. We propose a Likelihood Ratio based Model Selection method(LRMS)for the inference of granular model, which implements the fusion of minor models with the granular model. LRMS keeps the single-path calculating of Viterbi-based sequence labeling model, which avoid the violation of contextual connections. In empirical experiments on part-of-speech tagging, LRMS improves the accuracy in every transfer learning task, therefore, the effectiveness of LRMS in solving the under-adaptation problem is verified
%K 迁移学习
%K 似然比
%K 词性标注
%K 模型选择
%K &lt
%K br&gt
%K transfer learning
%K likelihood ratio
%K part-of-speech tagging
%K model selection
%U http://lxbwk.njournal.sdu.edu.cn/CN/10.6040/j.issn.1671-9352.5.2016.023