ICML 2018oral117 citations

Stein Points

Wilson Ye Chen, Lester Mackey, Jackson Gorham, Francois-Xavier Briol, Chris Oates

Abstract

An important task in computational statistics and machine learning is to approximate a posterior distribution $p(x)$ with an empirical measure supported on a set of representative points $\{x_i\}_{i=1}^n$. This paper focuses on methods where the selection of points is essentially deterministic, with an emphasis on achieving accurate approximation when $n$ is small. To this end, we present Stein Points. The idea is to exploit either a greedy or a conditional gradient method to iteratively minimise a kernel Stein discrepancy between the empirical measure and $p(x)$. Our empirical results demonstrate that Stein Points enable accurate approximation of the posterior at modest computational cost. In addition, theoretical results are provided to establish convergence of the method.

BibTeX
@InProceedings{pmlr-v80-chen18f,
  title = 	 {Stein Points},
  author =       {Chen, Wilson Ye and Mackey, Lester and Gorham, Jackson and Briol, Francois-Xavier and Oates, Chris},
  booktitle = 	 {Proceedings of the 35th International Conference on Machine Learning},
  pages = 	 {844--853},
  year = 	 {2018},
  editor = 	 {Dy, Jennifer and Krause, Andreas},
  volume = 	 {80},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {10--15 Jul},
  publisher =    {PMLR},
  pdf = 	 {http://proceedings.mlr.press/v80/chen18f/chen18f.pdf},
  url = 	 {https://proceedings.mlr.press/v80/chen18f.html},
  abstract = 	 {An important task in computational statistics and machine learning is to approximate a posterior distribution $p(x)$ with an empirical measure supported on a set of representative points $\{x_i\}_{i=1}^n$. This paper focuses on methods where the selection of points is essentially deterministic, with an emphasis on achieving accurate approximation when $n$ is small. To this end, we present Stein Points. The idea is to exploit either a greedy or a conditional gradient method to iteratively minimise a kernel Stein discrepancy between the empirical measure and $p(x)$. Our empirical results demonstrate that Stein Points enable accurate approximation of the posterior at modest computational cost. In addition, theoretical results are provided to establish convergence of the method.}
}
Stein Points · ICML 2018