AISTATS 2021poster31 citations

Optimal Quantisation of Probability Measures Using Maximum Mean Discrepancy

Onur Teymur, Jackson Gorham, Marina Riabiz, Chris Oates

Abstract

Several researchers have proposed minimisation of maximum mean discrepancy (MMD) as a method to quantise probability measures, i.e., to approximate a distribution by a representative point set. We consider sequential algorithms that greedily minimise MMD over a discrete candidate set. We propose a novel non-myopic algorithm and, in order to both improve statistical efficiency and reduce computational cost, we investigate a variant that applies this technique to a mini-batch of the candidate set at each iteration. When the candidate points are sampled from the target, the consistency of these new algorithms—and their mini-batch variants—is established. We demonstrate the algorithms on a range of important computational problems, including optimisation of nodes in Bayesian cubature and the thinning of Markov chain output.

BibTeX
@InProceedings{pmlr-v130-teymur21a,
  title = 	 { Optimal Quantisation of Probability Measures Using Maximum Mean Discrepancy },
  author =       {Teymur, Onur and Gorham, Jackson and Riabiz, Marina and Oates, Chris},
  booktitle = 	 {Proceedings of The 24th International Conference on Artificial Intelligence and Statistics},
  pages = 	 {1027--1035},
  year = 	 {2021},
  editor = 	 {Banerjee, Arindam and Fukumizu, Kenji},
  volume = 	 {130},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {13--15 Apr},
  publisher =    {PMLR},
  pdf = 	 {http://proceedings.mlr.press/v130/teymur21a/teymur21a.pdf},
  url = 	 {https://proceedings.mlr.press/v130/teymur21a.html},
  abstract = 	 { Several researchers have proposed minimisation of maximum mean discrepancy (MMD) as a method to quantise probability measures, i.e., to approximate a distribution by a representative point set. We consider sequential algorithms that greedily minimise MMD over a discrete candidate set. We propose a novel non-myopic algorithm and, in order to both improve statistical efficiency and reduce computational cost, we investigate a variant that applies this technique to a mini-batch of the candidate set at each iteration. When the candidate points are sampled from the target, the consistency of these new algorithms—and their mini-batch variants—is established. We demonstrate the algorithms on a range of important computational problems, including optimisation of nodes in Bayesian cubature and the thinning of Markov chain output. }
}
Optimal Quantisation of Probability Measures Using Maximum Mean Discrepancy · AISTATS 2021