Learning Convex Bodies is Hard

Learning Convex Bodies is Hard
复制标题

学习凸体很难

DOI:
--
复制
发表时间:
2009
期刊:
Annual Conference Computational Learning Theory
影响因子:
--
通讯作者:
Navin Goyal
Navin Goyal
中科院分区:
--
文献类型:
--
作者:
Luis Rademacher;Navin Goyal

文献摘要

被引文献

相似文献

我们证明了,在$RR^d$中学习凸体,给定来自凸体的随机样本,需要$2^{Omega(sqrt{d/eps})}$样本。通过学习凸体,我们的意思是找到一个与输入体最多有$eps$相对对称差的集合。为了证明下界,我们构造了一个很难学习的凸体族。我们构造这个族是非常简单的,并且基于纠错码。
We show that learning a convex body in $RR^d$, given random samples from the body, requires $2^{Omega(sqrt{d/eps})}$ samples. By learning a convex body we mean finding a set having at most $eps$ relative symmetric difference with the input body. To prove the lower bound we construct a hard to learn family of convex bodies. Our construction of this family is very simple and based on error correcting codes.