arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1909.02102math.OCcs.ITmath.ITstat.COstat.ML
收藏

加速信息梯度流

Accelerated Information Gradient flow

Yifei Wang, Wuchen Li

摘要:

我们提出了一个框架Nesterov的加速梯度流在概率空间设计有效的平均场马尔可夫链蒙特卡罗(MCMC)算法的贝叶斯逆问题。本文考虑了四种信息度量的例子,包括Fisher-Rao度量、Wasserstein-2度量、Kalman-Wasserstein度量和Stein度量。对于Fisher-Rao和Wasserstein-2度量,我们证明了加速梯度流的收敛性质。在实现中,我们提出了一个采样有效的离散时间算法的Wasserstein-2,Kalman-Wasserstein和Stein加速梯度流与重新启动技术。我们还制定了一个核带宽选择方法,学习密度的对数梯度从布朗运动样本。数值实验,包括贝叶斯逻辑斯谛回归和贝叶斯神经网络,表明所提出的方法相比,国家的最先进的算法的强度。

英文摘要:

We present a framework for Nesterov's accelerated gradient flows in probability space to design efficient mean-field Markov chain Monte Carlo (MCMC) algorithms for Bayesian inverse problems. Here four examples of information metrics are considered, including Fisher-Rao metric, Wasserstein-2 metric, Kalman-Wasserstein metric and Stein metric. For both Fisher-Rao and Wasserstein-2 metrics, we prove convergence properties of accelerated gradient flows. In implementations, we propose a sampling-efficient discrete-time algorithm for Wasserstein-2, Kalman-Wasserstein and Stein accelerated gradient flows with a restart technique. We also formulate a kernel bandwidth selection method, which learns the gradient of logarithm of density from Brownian-motion samples. Numerical experiments, including Bayesian logistic regression and Bayesian neural network, show the strength of the proposed methods compared with state-of-the-art algorithms.

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。

↑