Disentangled Representation for Causal Mediation Analysis
Proceedings of the AAAI Conference on Artificial Intelligence · 26.06.2023 · 37 (9) · 10666-10674
DOI: 10.1609/aaai.v37i9.26266 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 26.06.2023. Verfasst von Ziqi Xu und 5 weiteren Autor:innen. Die Arbeit wurde bisher 15-mal zitiert (Crossref).
Abstract (Originalsprache)
Estimating direct and indirect causal effects from observational data is crucial to understanding the causal mechanisms and predicting the behaviour under different interventions. Causal mediation analysis is a method that is often used to reveal direct and indirect effects. Deep learning shows promise in mediation analysis, but the current methods only assume latent confounders that affect treatment, mediator and outcome simultaneously, and fail to identify different types of latent confounders (e.g., confounders that only affect the mediator or outcome). Furthermore, current methods are based on the sequential ignorability assumption, which is not feasible for dealing with multiple types of latent confounders. This work aims to circumvent the sequential ignorability assumption and applies the piecemeal deconfounding assumption as an alternative. We propose the Disentangled Mediation Analysis Variational AutoEncoder (DMAVAE), which disentangles the representations of latent confounders into three types to accurately estimate the natural direct effect, natural indirect effect and total effect. Experimental results show that the proposed method outperforms existing methods and has strong generalisation ability. We further apply the method to a real-world dataset to show its potential application.
Zitieren
APA
Xu, Z., Cheng, D., Li, J., Liu, J., Liu, L., & Wang, K. (2023). Disentangled Representation for Causal Mediation Analysis. Proceedings of the AAAI Conference on Artificial Intelligence, 37(9), 10666-10674. https://doi.org/10.1609/aaai.v37i9.26266
Vancouver
Xu Z, Cheng D, Li J, Liu J, Liu L, Wang K. Disentangled Representation for Causal Mediation Analysis. Proceedings of the AAAI Conference on Artificial Intelligence. 2023;37(9):10666-10674. doi:10.1609/aaai.v37i9.26266
BibTeX
@article{Xu2023,
author = {Xu, Ziqi and Cheng, Debo and Li, Jiuyong and Liu, Jixue and Liu, Lin and Wang, Ke},
title = {Disentangled Representation for Causal Mediation Analysis},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2023},
volume = {37},
number = {9},
pages = {10666--10674},
doi = {10.1609/aaai.v37i9.26266}
}
EUVIMED