Improving Kernel-Based Nonasymptotic Simultaneous Confidence Bands

被引:0
|
作者
Csaji, Balazs Csanad [1 ,2 ]
Horvath, Balint [1 ,3 ]
机构
[1] Eotvos Lorand Res Network ELKH, Inst Comp Sci & Control SZTAKI, 13-17 Kende Utca, H-1111 Budapest, Hungary
[2] Eotvos Lorand Univ, Inst Math, 1-C Pazmany Petersetany, H-1117 Budapest, Hungary
[3] Budapest Univ Technol & Econ BME, Inst Math, 1 Egry Jozsef Utca, H-1111 Budapest, Hungary
来源
IFAC PAPERSONLINE | 2023年 / 56卷 / 02期
关键词
nonparametric methods; nonlinear system identification; statistical data analysis; estimation and filtering; convex optimization; randomized algorithms;
D O I
10.1016/j.ifacol.2023.10.1047
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
The paper studies the problem of constructing nonparametric simultaneous confidence bands with nonasymptotic and distribition-free guarantees. The target function is assumed to be band-limited and the approach is based on the theory of Paley-Wiener reproducing kernel Hilbert spaces. The starting point of the paper is a recently developed algorithm to which we propose three types of improvements. First, we relax the assumptions on the noises by replacing the symmetricity assumption with a weaker distributional invariance principle. Then, we propose a more efficient way to estimate the norm of the target function, and finally we enhance the construction of the confidence bands by tightening the constraints of the underlying convex optimization problems. The refinements are also illustrated through numerical experiments.
引用
收藏
页码:10357 / 10362
页数:6
相关论文
共 50 条