Invariant Ancestry Search

Phillip B Mogensen, Nikolaj Thams, Jonas Peters
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:15832-15857, 2022.

Abstract

Recently, methods have been proposed that exploit the invariance of prediction models with respect to changing environments to infer subsets of the causal parents of a response variable. If the environments influence only few of the underlying mechanisms, the subset identified by invariant causal prediction (ICP), for example, may be small, or even empty. We introduce the concept of minimal invariance and propose invariant ancestry search (IAS). In its population version, IAS outputs a set which contains only ancestors of the response and is a superset of the output of ICP. When applied to data, corresponding guarantees hold asymptotically if the underlying test for invariance has asymptotic level and power. We develop scalable algorithms and perform experiments on simulated and real data.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-mogensen22a, title = {Invariant Ancestry Search}, author = {Mogensen, Phillip B and Thams, Nikolaj and Peters, Jonas}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {15832--15857}, 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/mogensen22a/mogensen22a.pdf}, url = {https://proceedings.mlr.press/v162/mogensen22a.html}, abstract = {Recently, methods have been proposed that exploit the invariance of prediction models with respect to changing environments to infer subsets of the causal parents of a response variable. If the environments influence only few of the underlying mechanisms, the subset identified by invariant causal prediction (ICP), for example, may be small, or even empty. We introduce the concept of minimal invariance and propose invariant ancestry search (IAS). In its population version, IAS outputs a set which contains only ancestors of the response and is a superset of the output of ICP. When applied to data, corresponding guarantees hold asymptotically if the underlying test for invariance has asymptotic level and power. We develop scalable algorithms and perform experiments on simulated and real data.} }
Endnote
%0 Conference Paper %T Invariant Ancestry Search %A Phillip B Mogensen %A Nikolaj Thams %A Jonas Peters %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-mogensen22a %I PMLR %P 15832--15857 %U https://proceedings.mlr.press/v162/mogensen22a.html %V 162 %X Recently, methods have been proposed that exploit the invariance of prediction models with respect to changing environments to infer subsets of the causal parents of a response variable. If the environments influence only few of the underlying mechanisms, the subset identified by invariant causal prediction (ICP), for example, may be small, or even empty. We introduce the concept of minimal invariance and propose invariant ancestry search (IAS). In its population version, IAS outputs a set which contains only ancestors of the response and is a superset of the output of ICP. When applied to data, corresponding guarantees hold asymptotically if the underlying test for invariance has asymptotic level and power. We develop scalable algorithms and perform experiments on simulated and real data.
APA
Mogensen, P.B., Thams, N. & Peters, J.. (2022). Invariant Ancestry Search. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:15832-15857 Available from https://proceedings.mlr.press/v162/mogensen22a.html.

Related Material