Adaptive optimal transport

Adaptive optimal transport
复制标题

自适应最优传输

DOI:
10.1093/imaiai/iaz008
复制
发表时间:
2019
期刊:
Information and Inference: A Journal of the IMA
影响因子:
--
通讯作者:
Tabak, Esteban G.
Tabak, Esteban G.
中科院分区:
--
文献类型:
--
作者:
Essid, Montacer;Laefer, Debra F.;Tabak, Esteban G.

文献摘要

参考文献

被引文献

相似文献

一个自适应的,对抗性的方法是开发两个distributionand之间的最优运输问题,只知道通过一个有限的一组独立的samplesand。该方法自动创建适应数据的特征,从而避免依赖于数据基础分布的优先知识。具体来说,而不是一个离散的逐点分配,新的程序寻求一个最佳的mapdefined为所有,最大限度地减少Kullback-Leibler分歧之间的和目标。相对熵被赋予了一个基于样本的变分特征,从而创造了一个对抗性的设置:当一个玩家试图将一个分布推向另一个时,第二个玩家开发了专注于两个分布无法匹配的区域的特征。该程序解决了局部问题,寻求连续的,中间分布之间的最佳转移和。因此,可以通过组合用于每个局部问题的简单映射来构建任意复杂度的映射。位移插值用于保证全局局部最优。该过程是通过在一个和两个维度的合成的例子说明。
An adaptive, adversarial methodology is developed for the optimal transport problem between two distributionsand, known only through a finite set of independent samplesand. The methodology automatically creates features that adapt to the data, thus avoiding reliance ona prioriknowledge of the distributions underlying the data. Specifically, instead of a discrete point-by-point assignment, the new procedure seeks an optimal mapdefined for all, minimizing the Kullback–Leibler divergence betweenand the target. The relative entropy is given a sample-based, variational characterization, thereby creating an adversarial setting: as one player seeks to push forward one distribution to the other, the second player develops features that focus on those areas where the two distributions fail to match. The procedure solves local problems that seek the optimal transfer between consecutive, intermediate distributions betweenand. As a result, maps of arbitrary complexity can be built by composing the simple maps used for each local problem. Displaced interpolation is used to guarantee global from local optimality. The procedure is illustrated through synthetic examples in one and two dimensions.
DOI: 10.1093/imaiai/iax023
发表时间: 2018
期刊: Information and Inference: A Journal of the IMA
影响因子: --
作者:
E. Tabak;G. Trigila
通讯作者: G. Trigila
DOI: --
发表时间: 2017
期刊: Optimization
影响因子: 2.2
作者:
M. Pavon
通讯作者: M. Pavon
极小极大算法
DOI: --
发表时间: 1989
期刊:
影响因子: --
作者:
M. Minasi
通讯作者: M. Minasi
用于密度估计的内核数据压缩
DOI: --
发表时间: 2006
期刊:
影响因子: --
作者:
Atsuyuki;Kogure;Masahiko;Sagae
通讯作者: Sagae
通过最大似然密度比估计来近似互信息
DOI: --
发表时间: 2008
期刊:
影响因子: --
作者:
Akimichi Takemura;Taiji Suzuki;Taiji Suzuki;杉山将;Jun Sese;鈴木大慈
通讯作者: 鈴木大慈