中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 自动化研究所
  3. 中国科学院自动化研究所
  4. 多模态人工智能系统全国重点实验室
Large-scale continual learning for ancient Chinese character recognition

文献类型：期刊论文

作者Xu, Yue1,2; Zhang, Xu-Yao1,2; Zhang, Zhaoxiang1,2; Liu, Cheng-Lin1,2
刊名PATTERN RECOGNITION
出版日期2024-06-01
卷号150页码:15
关键词Continual learning Class-incremental learning Convolutional prototype network Character recognition Ancient Chinese characters
ISSN号0031-3203
DOI10.1016/j.patcog.2024.110283
通讯作者Zhang, Xu-Yao(xyz@nlpr.ia.ac.cn)
英文摘要Ancient Chinese character recognition is a challenging problem in the field of pattern recognition. It is difficult to collect all character classes during the training stage due to the numerous classes of ancient Chinese characters and the likelihood of discovering new characters over time. A solution to address this problem is continual learning. However, most continual learning methods are not well -suited for large-scale applications, making them insufficient for solving the problem of ancient Chinese character recognition. Although saving raw data for old classes is a good approach for continual learning to address large-scale problems, it is often infeasible due to the lack of data accessibility in reality. To solve these problems, we propose a large-scale continual learning framework based on the convolutional prototype network (CPN), which does not save raw data for old classes. In this paper, several basic strategies have been proposed for the initial training stage to enhance the feature extraction ability and robustness of the network, which can improve the performance of the model in continual learning. In addition, we propose two practical methods in varying feature space (parameters of feature extractor are changeable) and fixed feature space (parameters of feature extractor are fixed), which enable the model to carry out large-scale continual learning. The proposed method does not save the raw data of old classes and enables simultaneous classification of all existing classes without knowing the incremental batch number. Experiments on the CASIA-AHCDB dataset with 5000 character classes demonstrate the effectiveness and superiority of the proposed method.
WOS研究方向Computer Science ; Engineering
语种英语
WOS记录号WOS:001176417300001
出版者ELSEVIER SCI LTD
源URL[http://ir.ia.ac.cn/handle/173211/56972]  
专题多模态人工智能系统全国重点实验室
通讯作者Zhang, Xu-Yao
作者单位1.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing, Peoples R China
2.Chinese Acad Sci, State Key Lab Multimodal Artificial Intelligence S, Inst Automat, Beijing, Peoples R China
推荐引用方式
GB/T 7714		 Xu, Yue,Zhang, Xu-Yao,Zhang, Zhaoxiang,et al. Large-scale continual learning for ancient Chinese character recognition[J]. PATTERN RECOGNITION,2024,150:15.
APA Xu, Yue,Zhang, Xu-Yao,Zhang, Zhaoxiang,&Liu, Cheng-Lin.(2024).Large-scale continual learning for ancient Chinese character recognition.PATTERN RECOGNITION,150,15.
MLA Xu, Yue,et al."Large-scale continual learning for ancient Chinese character recognition".PATTERN RECOGNITION 150(2024):15.

入库方式： OAI收割

来源：自动化研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。