Pre-training and/or transfer learning for sequence taggers

作者: Ruhi Sarikaya , Minwoo Jeong , Young-Bum Kim

DOI:

关键词:

摘要: Systems and methods for pre-training a sequence tagger with unlabeled data, such as hidden layered conditional random field model are provided. Additionally, systems transfer learning Accordingly, the build more accurate, reliable, and/or efficient taggers than previously utilized that not pre-trained data capable of learning/training.

参考文章(22)
Pavel P. Kuksa, Yanjun Qi, Semi-supervised Bio-named Entity Recognition with Word-Codebook Learning. siam international conference on data mining. pp. 25- 36 ,(2010)
Li Deng, Dong Yu, Abdel-rahman Samir Abdel-rahman Mohamed, Full-sequence training of deep structures for speech recognition ,(2011)
ZhongFei (Mark) Zhang, Zhen Guo, Semi-supervised learning based on semiparametric regularization siam international conference on data mining. pp. 132- 142 ,(2009)
Ruhi Sarikaya, Geoffrey E. Hinton, Anoop Deoras, Application of Deep Belief Networks for natural language understanding IEEE Transactions on Audio, Speech, and Language Processing. ,vol. 22, pp. 778- 784 ,(2014) , 10.1109/TASLP.2014.2303296
Asela J. Gunawardana, Hidden conditional random field models for phonetic classification and speech recognition Journal of the Acoustical Society of America. ,vol. 127, pp. 3294- ,(2004) , 10.1121/1.3432308
Antoine Vinel, Trinh Minh Tri Do, Thierry Artieres, Joint Optimization of Hidden Conditional Random Fields and Non Linear Feature Extraction international conference on document analysis and recognition. pp. 513- 517 ,(2011) , 10.1109/ICDAR.2011.109
Ruhi Sarikaya, Asli Celikyilmaz, Dilek Hakkani-Tur, Gokhan Tur, Semi-Supervised Semantic Tagging of Conversational Understanding using Markov Topic Regression meeting of the association for computational linguistics. ,vol. 1, pp. 914- 923 ,(2013)
Ronan Collobert, Jason Weston, A unified architecture for natural language processing Proceedings of the 25th international conference on Machine learning - ICML '08. pp. 160- 167 ,(2008) , 10.1145/1390156.1390177