The recovery of ridge functions on the hypercube suffers from the curse of dimensionality

The recovery of ridge functions on the hypercube suffers from the curse of dimensionality
复制标题

超立方体上岭函数的恢复遭受维数灾难

DOI:
10.1016/j.jco.2020.101521
复制
发表时间:
2021
期刊:
J. Complex.
影响因子:
--
通讯作者:
S. Mayer
S. Mayer
中科院分区:
--
文献类型:
--
作者:
B. Doerr;S. Mayer

文献摘要

参考文献

被引文献

相似文献

多元脊函数是形式为f (x)= g (A T x)的函数,其中g是单变量的,A∈R d。我们表明,当在L∞范数中测量恢复误差时,定义在超立方体[- 1,1]d上具有lipschitz -正则轮廓g的未知脊函数的恢复遭受维数诅咒,即使我们允许随机算法。如果a的有限数量的分量比其他分量大得多,那么维数的诅咒就不存在,只要轮廓g足够规则,这个问题就很容易处理。
A multivariate ridge function is a function of the form f (x)= g (a T x), where g is univariate and a∈ R d. We show that the recovery of an unknown ridge function defined on the hypercube [− 1, 1] d with Lipschitz-regular profile g suffers from the curse of dimensionality when the recovery error is measured in the L∞-norm, even if we allow randomized algorithms. If a limited number of components of a is substantially larger than the others, then the curse of dimensionality is not present and the problem is weakly tractable, provided the profile g is sufficiently regular.
DOI: 10.1007/s00365-015-9282-6
发表时间: 2014-02
影响因子: 2.7
作者:
E. Novak;Daniel Rudolf
通讯作者: E. Novak;Daniel Rudolf
DOI: 10.1007/s00365-011-9147-6
发表时间: 2011
影响因子: 2.7
作者:
A. Cohen;I. Daubechies;R. DeVore;G. Kerkyacharian;D. Picard
通讯作者: D. Picard
DOI: 10.1017/cbo9781316408124.011
发表时间: 2015
期刊: Statistical Analysis and Data Mining: The ASA Data Science Journal
影响因子: --
作者:
A. Pinkus
通讯作者: A. Pinkus
DOI: 10.1007/s00365-014-9267-x
发表时间: 2015
影响因子: 2.7
作者:
S. Mayer;T. Ullrich;J. Vybiral
通讯作者: J. Vybiral
组合数学中的概率方法
DOI: 10.1007/978-3-0348-9078-6_132
发表时间: 1974
影响因子: 0.8
作者:
J. Spencer
通讯作者: J. Spencer