An Asymptotic Test for Conditional Independence using Analytic Kernel Embeddings

Meyer Scetbon, Laurent Meunier, Yaniv Romano
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:19328-19346, 2022.

Abstract

We propose a new conditional dependence measure and a statistical test for conditional independence. The measure is based on the difference between analytic kernel embeddings of two well-suited distributions evaluated at a finite set of locations. We obtain its asymptotic distribution under the null hypothesis of conditional independence and design a consistent statistical test from it. We conduct a series of experiments showing that our new test outperforms state-of-the-art methods both in terms of type-I and type-II errors even in the high dimensional setting.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-scetbon22a, title = {An Asymptotic Test for Conditional Independence using Analytic Kernel Embeddings}, author = {Scetbon, Meyer and Meunier, Laurent and Romano, Yaniv}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {19328--19346}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/scetbon22a/scetbon22a.pdf}, url = {https://proceedings.mlr.press/v162/scetbon22a.html}, abstract = {We propose a new conditional dependence measure and a statistical test for conditional independence. The measure is based on the difference between analytic kernel embeddings of two well-suited distributions evaluated at a finite set of locations. We obtain its asymptotic distribution under the null hypothesis of conditional independence and design a consistent statistical test from it. We conduct a series of experiments showing that our new test outperforms state-of-the-art methods both in terms of type-I and type-II errors even in the high dimensional setting.} }
Endnote
%0 Conference Paper %T An Asymptotic Test for Conditional Independence using Analytic Kernel Embeddings %A Meyer Scetbon %A Laurent Meunier %A Yaniv Romano %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-scetbon22a %I PMLR %P 19328--19346 %U https://proceedings.mlr.press/v162/scetbon22a.html %V 162 %X We propose a new conditional dependence measure and a statistical test for conditional independence. The measure is based on the difference between analytic kernel embeddings of two well-suited distributions evaluated at a finite set of locations. We obtain its asymptotic distribution under the null hypothesis of conditional independence and design a consistent statistical test from it. We conduct a series of experiments showing that our new test outperforms state-of-the-art methods both in terms of type-I and type-II errors even in the high dimensional setting.
APA
Scetbon, M., Meunier, L. & Romano, Y.. (2022). An Asymptotic Test for Conditional Independence using Analytic Kernel Embeddings. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:19328-19346 Available from https://proceedings.mlr.press/v162/scetbon22a.html.

Related Material