Generalized Independent Noise Condition for Estimating Linear Non-Gaussian Latent Variable Graphs

10/10/2020
βˆ™
by   Feng Xie, et al.
βˆ™
4
βˆ™

Causal discovery aims to recover causal structures or models underlying the observed data. Despite its success in certain domains, most existing methods focus on causal relations between observed variables, while in many scenarios the observed ones may not be the underlying causal variables (e.g., image pixels), but are generated by latent causal variables or confounders that are causally related. To this end, in this paper, we consider Linear, Non-Gaussian Latent variable Models (LiNGLaMs), in which latent confounders are also causally related, and propose a Generalized Independent Noise (GIN) condition to estimate such latent variable graphs. Specifically, for two observed random vectors 𝐘 and 𝐙, GIN holds if and only if Ο‰^⊺𝐘 and 𝐙 are statistically independent, where Ο‰ is a parameter vector characterized from the cross-covariance between 𝐘 and 𝐙. From the graphical view, roughly speaking, GIN implies that causally earlier latent common causes of variables in 𝐘 d-separate 𝐘 from 𝐙. Interestingly, we find that the independent noise condition, i.e., if there is no confounder, causes are independent from the error of regressing the effect on the causes, can be seen as a special case of GIN. Moreover, we show that GIN helps locate latent variables and identify their causal structure, including causal directions. We further develop a recursive learning algorithm to achieve these goals. Experimental results on synthetic and real-world data demonstrate the effectiveness of our method.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
βˆ™ 03/15/2012

Invariant Gaussian Process Latent Variable Models and Application in Causal Discovery

In nonlinear latent variable models or dynamic models, if we consider th...
research
βˆ™ 09/19/2020

Causal Discovery with Multi-Domain LiNGAM for Latent Factors

Discovering causal structures among latent factors from observed data is...
research
βˆ™ 05/04/2023

Learning to Recover Causal Relationship from Indefinite Data in the Presence of Latent Confounders

In Causal Discovery with latent variables, We define two data paradigms:...
research
βˆ™ 02/12/2021

Do-calculus enables causal reasoning with latent variable models

Latent variable models (LVMs) are probabilistic models where some of the...
research
βˆ™ 02/12/2021

A Critical Look At The Identifiability of Causal Effects with Deep Latent Variable Models

Using deep latent variable models in causal inference has attracted cons...
research
βˆ™ 05/26/2023

Causal Component Analysis

Independent Component Analysis (ICA) aims to recover independent latent ...
research
βˆ™ 07/26/2020

MIMIC modelling with instrumental variables: A 2SLS-MIMIC approach

Multiple Indicators Multiple Causes (MIMIC) models are type of structura...

Please sign up or login with your details

Forgot password? Click here to reset