Broad Learning System Based on Maximum Correntropy Criterion
文献类型：期刊论文
| 作者 | Zheng, Yunfei1; Chen, Badong1; Wang, Shiyuan2; Wang, Weiqun3
|
| 刊名 | IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
 |
| 出版日期 | 2021-07-01 |
| 卷号 | 32期号:7页码:3083-3097 |
| ISSN号 | 2162-237X |
| 关键词 | Learning systems Robustness Standards Optimization Training Perturbation methods Mean square error methods Broad learning system (BLS) incremental learning algorithms maximum correntropy criterion (MCC) regression and classification |
| DOI | 10.1109/TNNLS.2020.3009417 |
| 通讯作者 | Chen, Badong(chenbd@mail.xjtu.edu.cn) |
| 英文摘要 | As an effective and efficient discriminative learning method, broad learning system (BLS) has received increasing attention due to its outstanding performance in various regression and classification problems. However, the standard BLS is derived under the minimum mean square error (MMSE) criterion, which is, of course, not always a good choice due to its sensitivity to outliers. To enhance the robustness of BLS, we propose in this work to adopt the maximum correntropy criterion (MCC) to train the output weights, obtaining a correntropy-based BLS (C-BLS). Due to the inherent superiorities of MCC, the proposed C-BLS is expected to achieve excellent robustness to outliers while maintaining the original performance of the standard BLS in the Gaussian or noise-free environment. In addition, three alternative incremental learning algorithms, derived from a weighted regularized least-squares solution rather than pseudoinverse formula, for C-BLS are developed. With the incremental learning algorithms, the system can be updated quickly without the entire retraining process from the beginning when some new samples arrive or the network deems to be expanded. Experiments on various regression and classification data sets are reported to demonstrate the desirable performance of the new methods. |
| WOS关键词 | FIXED-POINT ALGORITHM ; UNIVERSAL APPROXIMATION ; NEURAL-NETWORKS ; CONVERGENCE |
| 资助项目 | National Natural Science Foundation of China[91648208] ; National Natural Science Foundation of China[61976175] ; National Natural Science Foundation-Shenzhen Joint Research Program[U1613219] ; Key Project of Natural Science Basic Research Plan in Shaanxi Province of China[2019JZ-05] |
| WOS研究方向 | Computer Science ; Engineering |
| 语种 | 英语 |
| 出版者 | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC |
| WOS记录号 | WOS:000670541500023 |
| 资助机构 | National Natural Science Foundation of China ; National Natural Science Foundation-Shenzhen Joint Research Program ; Key Project of Natural Science Basic Research Plan in Shaanxi Province of China |
| 源URL | [http://ir.ia.ac.cn/handle/173211/45252]  |
| 专题 | 自动化研究所_复杂系统管理与控制国家重点实验室_先进机器人控制团队 |
| 通讯作者 | Chen, Badong |
| 作者单位 | 1.Xi An Jiao Tong Univ, Inst Artificial Intelligence & Robot, Xian 710049, Peoples R China 2.Southwest Univ, Coll Elect & Informat Engn, Chongqing 400715, Peoples R China 3.Chinese Acad Sci, Inst Automat, State Key Lab Management & Control Complex Syst, Beijing 100190, Peoples R China |
| 推荐引用方式 GB/T 7714 | Zheng, Yunfei,Chen, Badong,Wang, Shiyuan,et al. Broad Learning System Based on Maximum Correntropy Criterion[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS,2021,32(7):3083-3097. |
| APA | Zheng, Yunfei,Chen, Badong,Wang, Shiyuan,&Wang, Weiqun.(2021).Broad Learning System Based on Maximum Correntropy Criterion.IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS,32(7),3083-3097. |
| MLA | Zheng, Yunfei,et al."Broad Learning System Based on Maximum Correntropy Criterion".IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 32.7(2021):3083-3097. |
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。