DeepAI AI Chat
Log In Sign Up

Moment-Matching Graph-Networks for Causal Inference

by   Michael Park, et al.

In this note we explore a fully unsupervised deep-learning framework for simulating non-linear structural equation models from observational training data. The main contribution of this note is an architecture for applying moment-matching loss functions to the edges of a causal Bayesian graph, resulting in a generative conditional-moment-matching graph-neural-network. This framework thus enables automated sampling of latent space conditional probability distributions for various graphical interventions, and is capable of generating out-of-sample interventional probabilities that are often faithful to the ground truth distributions well beyond the range contained in the training set. These methods could in principle be used in conjunction with any existing autoencoder that produces a latent space representation containing causal graph structures.


page 1

page 2

page 3

page 4


Axiomatization of Interventional Probability Distributions

Causal intervention is an essential tool in causal inference. It is axio...

Learning Causal Relationships from Conditional Moment Conditions by Importance Weighting

We consider learning causal relationships under conditional moment condi...

Approaching an unknown communication system by latent space exploration and causal inference

This paper proposes a methodology for discovering meaningful properties ...

Spectral Representation Learning for Conditional Moment Models

Many problems in causal inference and economics can be formulated in the...

CausalGAN: Learning Causal Implicit Generative Models with Adversarial Training

We propose an adversarial training procedure for learning a causal impli...

Identification of Latent Variables From Graphical Model Residuals

Graph-based causal discovery methods aim to capture conditional independ...