摘要
Modeling relations between local optimum parameter vectors in multitask networks has attracted much attention over the last years. This work considers a distributed optimization problem for parameter vectors with a jointly sparse structure among nodes, that is, the parameter vectors share the same support set. By introducing an l(infinity,1)-norm penalty at each node, and using a proximal gradient method to minimize the regularized cost, we devise a proximal multitask diffusion LMS algorithm which promotes the joint-sparsity to enhance the estimation performance. Analyses are provided to ensure the stability. Simulation results are presented to highlight the performance.
| 源语言 | 英语 |
|---|---|
| 页(从-至) | 5900-5904 |
| 页数 | 5 |
| 期刊 | 2020 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING |
| DOI | |
| 出版状态 | 已出版 - 4 5月 2020 |
| 活动 | IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP) - New Orleans,LA, New Orleans, 老挝人民民主共和国 期限: 5 3月 2017 → 9 3月 2017 |
基金资助
The work of Jie Chen was supported in part by NSFC grants 61671382 and 61811530283. The work of Jingdong Chen was supported in part by NSFC grants 61761146001 and 61811530283. The work of C. Richard was funded in part by ANR under grant ANR-19-CE48-0002 and by the CoopIn-tEER program CNRS-NSFC (DIALOG project). Corresponding author: Jie Chen.
关键词
- Distributed optimization
- diffusion strategy
- proximal operator
- joint sparsity
- l(infinity,1)-norm regularization
收录
- CPCI-S
学术指纹
探究 'PROXIMAL MULTITASK LEARNING OVER DISTRIBUTED NETWORKS WITH JOINTLY SPARSE STRUCTURE' 的科研主题。它们共同构成独一无二的学术指纹。引用此
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver