| 摘要: |
| 本文研究了基于函数型输入和l1-正则化的最小二乘回归问题的推广性能.利用基于Rademacher平均的分析技术,获得了学习速度的估计,推广了已有的欧式空间有限维输入结果. |
| 关键词: 回归 函数型数据 l1-正则化 Rademacher平均 |
| DOI： |
| 分类号:O212.1 |
| 基金项目:Supported partially by National Natural Science Foundation of China (61105051). |
|
| ON THE CONVERGENCE RATE OF COEFFICIENT-BASED REGULARIZED REGRESSION FOR FUNCTIONAL DATA |
|
TAO Yan-fang1, TANG Yi2
|
|
1.Dept. of Basis, Changjiang Professional College, Wuhan 430074, China;2.School of Math. and Computer Sci., Yunnan University of Nationalities, Kunming 650031, China
|
| Abstract: |
| This paper investigates the generalization performance of least square regression with functional data and l1-regularizer. The estimate of learning rate is established by Rademacher average technique. The theoretical result is a natural extension for coefficient-based regularized regression when input space is a subset of infinite-dimensional Euclidean space. |
| Key words: regression functional data l1-regularizer Rademacher average |
