Rapid Convergence of the Unadjusted Langevin Algorithm: Log-Sobolev Suffices

Rapid Convergence of the Unadjusted Langevin Algorithm: Log-Sobolev Suffices
复制标题

DOI:
--
复制
发表时间:
2019-03
期刊:
ArXiv
影响因子:
--
通讯作者:
S. Vempala;Andre Wibisono
S. Vempala;Andre Wibisono
中科院分区:
其他
文献类型:
--
作者:
S. Vempala;Andre Wibisono

文献摘要

相似文献

在假设目标分布$e^-f}$满足Log-Soblev不等式且$f$的Hessian有界的条件下,证明了未调整的朗之万算法的收敛保证。特别地,$f$不需要是凸的,也不需要有更高的导数有界。
We prove a convergence guarantee on the unadjusted Langevin algorithm for sampling assuming only that the target distribution $e^{-f}$ satisfies a log-Sobolev inequality and the Hessian of $f$ is bounded. In particular, $f$ is not required to be convex or have higher derivatives bounded.