A4 Article in conference proceedings
Identifying Causal Effects via Context-specific Independence Relations (2019)


Tikka, Santtu; Hyttinen, Antti; Karvanen, Juha (2019). Identifying Causal Effects via Context-specific Independence Relations. In Wallach, H.; Larochelle, H.; Beygelzimer, A.; d'Alché-Buc, F.; Fox, E.; Garnett, R. (Eds.) NeurIPS 2019 : Proceedings of the 33rd Conference on Neural Information Processing Systems, 32. Neural Information Processing Systems Foundation, Inc.. https://papers.nips.cc/paper/8547-identifying-causal-effects-via-context-specific-independence-relations


JYU authors or editors


Publication details

All authors or editors: Tikka, Santtu; Hyttinen, Antti; Karvanen, Juha

Parent publication: NeurIPS 2019 : Proceedings of the 33rd Conference on Neural Information Processing Systems

Parent publication editors: Wallach, H.; Larochelle, H.; Beygelzimer, A.; d'Alché-Buc, F.; Fox, E.; Garnett, R.

Conference:

Advances in neural information processing systems

Place and date of conference: Vancouver, Canada, 8.-14.12.2019

ISSN: 1049-5258

Publication year: 2019

Number in series: 32

Publisher: Neural Information Processing Systems Foundation, Inc.

Publication country: United States

Publication language: English

Persistent website address: https://papers.nips.cc/paper/8547-identifying-causal-effects-via-context-specific-independence-relations

Open Access: Publication published in an open access channel

Publication is parallel published (JYX): https://jyx.jyu.fi/handle/123456789/67291


Abstract

Causal effect identification considers whether an interventional probability distribution can be uniquely determined from a passively observed distribution in a given causal structure. If the generating system induces context-specific independence (CSI) relations, the existing identification procedures and criteria based on do-calculus are inherently incomplete. We show that deciding causal effect non-identifiability is NP-hard in the presence of CSIs. Motivated by this, we design a calculus and an automated search procedure for identifying causal effects in the presence of CSIs. The approach is provably sound and it includes standard do-calculus as a special case. With the approach we can obtain identifying formulas that were unobtainable previously, and demonstrate that a small number of CSI-relations may be sufficient to turn a previously non-identifiable instance to identifiable.


Keywords: causality

Free keywords: causal effect identification; context-specific independence relations


Contributing organizations


Related projects


Ministry reporting: Yes

Reporting Year: 2019

JUFO rating: 2


Last updated on 2020-18-08 at 13:20