There has been recent interest in developing scalable Bayesian sampling methods for big-data analysis, such as stochastic gradient MCMC (SG-MCMC) and Stein variational gradient descent (SVGD). A standard SG-MCMC algorithm simulates samples from a discrete-time Markov chain to approximate a target distribution, thus samples could be highly correlated, an undesired property for SG-MCMC. In contrary, SVGD directly optimizes a set of particles to approximate a target distribution, and thus is able to obtain good approximate with relatively much fewer samples. In this paper, we propose a principle particle-optimization framework based on Wasserstein gradient flows to unify SG-MCMC and SVGD, and to allow new algorithms to be developed. Our framework interprets SG-MCMC as particle optimization, revealing strong connections between SG-MCMC and SVGD. The key component of our framework is several particle-approximate techniques to efficiently solve the original partial differential equations on the space of probability measures. Extensive experiments on both synthetic data and deep neural networks demonstrate the effectiveness and efficiency of our framework for scalable Bayesian sampling.

本文针对大数据分析，提出了一种基于Wasserstein梯度流的粒子优化框架，用于统一随机梯度MCMC和Stein变分梯度下降算法，并能够更有效地解决概率测度空间上的挑战。实验结果表明，该框架能够提高贝叶斯抽样的效率和可伸缩性。

面向可扩展贝叶斯采样的统一粒子优化框架