Accelerating Graph-Based Dependency Parsing with Lock-Free Parallel Perceptron

摘要：

　　Dependency parsing is an important NLP task.A popular approach for dependency parsing is structured perceptron.Still,graphbased dependency parsing has the time complexity of O(n3),and it suffers from slow training.To deal with this problem,we propose a parallel algorithm called parallel perceptron.The parallel algorithm can make full use of a multi-core computer which saves a lot of training time.Based on experiments we observe that dependency parsing with parallel perceptron can achieve 8-fold faster training speed than traditional structured perceptron methods when using 10 threads,and with no loss at all in accuracy.

关键词： Dependency parsing Lock-free Structured perceptron

作者: Shuming Ma Xu Sun Yi Zhang Bingzhen Wei

作者单位: MOE Key Lab of Computational Linguistics,School of EECS,Peking University,Beijing,China

会议类型: 国际会议

会议名称: 2018自然语言处理与中文计算国际会议(NLPCC2018)

会议地点: 呼和浩特

会议语种:英文

页码: 260-268

在线出版日期: 2018-08-26（万方平台首次上网日期，不代表论文的发表时间）