固定维度下通过反向扩散实现多项式查询复杂度的多模态分布采样
Sampling from multi-modal distributions with polynomial query complexity in fixed dimension via reverse diffusion
AI总结:
本文提出一种固定维度下基于反向扩散的多模态分布采样算法,通过自归一化蒙特卡洛估计得分函数,实现多项式查询复杂度,避免亚稳态且无需模态先验知识。
摘要:
即使在低维空间中,从多模态分布中采样也具有挑战性。我们首次提出了一种针对广泛分布类别(包括所有高斯混合)的采样算法,在固定维度的假设下,其查询复杂度关于控制多模态性的参数呈多项式增长。我们的采样算法模拟时间反转的扩散过程,使用中间得分函数的自归一化蒙特卡洛估计器。与以往的工作不同,该算法避免了亚稳态问题,无需预先知道模态位置,并放宽了著名的对数光滑性假设,该假设此前排除了通用高斯混合。
英文摘要:
Even in low dimensions, sampling from multi-modal distributions is challenging. We provide the first sampling algorithm for a broad class of distributions -- including all Gaussian mixtures -- with a query complexity that is polynomial in the parameters governing multi-modality, assuming fixed dimension. Our sampling algorithm simulates a time-reversed diffusion process, using a self-normalized Monte Carlo estimator of the intermediate score functions. Unlike previous works, it avoids metastability, requires no prior knowledge of the mode locations, and relaxes the well-known log-smoothness assumption which excluded general Gaussian mixtures so far.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。