Paper Info
Title
Semi-Supervised Transfer Learning for Convolutional Neural Network Based Chinese Character Recognition
Abstract
Although transfer learning has aroused researchers' great interest, how to utilize the unlabeled data is still an open and important problem in this area. We propose a novel semi-supervised transfer learning (STL) method by incorporating Multi-Kernel Maximum Mean Discrepancy (MK-MMD) loss into the traditional fine-tuned Convolutional Neural Network (CNN) transfer learning framework for Chinese character recognition. The proposed method includes three steps. First, a CNN model is trained by massive labeled samples in the source domain. Then the CNN model is fine-tuned by a few labeled samples in the target domain. Finally, the CNN model is trained with both a large number of unlabeled samples and the limited labeled samples in the target domain to minimize the MK-MMD loss. Experiments investigate detailed configurations and parameters of the proposed STL method with several frequently used CNN structures including AlexNet, GoogLeNet, and ResNet. Experimental results on practical Chinese character transfer learning tasks, such as Dunhuang historical Chinese character recognition, indicate that the proposed method can significantly improve recognition accuracy in the target domain.
Year
DOI
Venue
2017
10.1109/ICDAR.2017.79
2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR)
Keywords
Field
DocType
semi-supervised learning,transfer learning,convolutional neural network,Chinese character recognition
Kernel (linear algebra),Semi-supervised learning,Task analysis,Pattern recognition,Character recognition,Convolutional neural network,Computer science,Transfer of learning,Feature extraction,Probability distribution,Artificial intelligence
Conference
Volume
ISSN
ISBN
01
1520-5363
978-1-5386-3587-2
Citations 
PageRank 
References 
0
0.34
0
Authors
4
Name
Order
Citations
PageRank
Yejun Tang101.35
Bing Wu222.43
Liangrui Peng38017.67
Changsong Liu435836.20