Unsupervised Causal Abstractions Discovery

Théo Saulus; Simon Lacoste-Julien; Dhanya Sridhar

cs.LGJun 17, 2026

Unsupervised Causal Abstractions Discovery

Authors: Théo Saulus, Simon Lacoste-Julien, Dhanya Sridhar

Organizations: Mila - Quebec AI Institute · Université de Montréal · Canada CIFAR AI Chair

Abstract

Causal abstractions formalize when a high-level structural causal model (SCM) captures the interventional behavior of a lower-level SCM. Existing applications of this notion largely follow a hypothesis-testing paradigm: an expert proposes a candidate high-level model and then evaluates if the low-level system implements it. We study the complementary problem of learning a high-level model directly from low-level measurements. Our contributions leverage hypotheses from low-rank causal discovery, and can be summarized as follows: (1) we show that observations generated by a low-rank graph induce latents that form a causal abstraction, (2) we provide identifiability results about these latents, and (3) we propose a practical objective to learn this high-level SCM.

Explore similar work

Causal Learning with the Invariance Principle
May 13, 2026Francesco Montagna, Francesco LocatelloStructural Causal Model Causal Graph
Validating Causal Abstraction Metrics on Simulated Complex Systems
Jun 30, 2026Maxime Méloux, Tiago Pimentel, François Portet +1Causal Reasoning
Coarsening Linear Non-Gaussian Causal Models with Cycles
May 11, 2026Francisco Madaleno, Francisco C Pereira, Alex MarkhamCausal Representation Learning Structural Causal Model

May 13, 2026 · stat.MLJ/K move · Enter open · S save

Causal Learning with the Invariance Principle

Francesco Montagna, Francesco Locatello

Institute of Science and Technology Austria

Causal discovery, the problem of inferring the direction of causality, is generally ill-posed. We use the language of structural causal models (SCM) to show that assuming that the causal relations are acyclic and invariant across multiple environments (e.g., the way minimum wage affects employment rate is stable across different geographical regions), \textit{only} two auxiliary environments are sufficient to infer the causal graph for arbitrary nonlinear mechanisms. Moreover, we demonstrate that this implies identifiability of the SCM functional mechanisms: as a corollary, we show that \textit{two} auxiliary environments are sufficient to guarantee correct counterfactual inference. We empirically support our theoretical results on synthetic data.

Structural Causal Model Causal Graph Causal Reasoning Identifiability Counterfactual Reasoning Principles

Open paper Ask about it PDF

Abstract

Explore similar work

Causal Learning with the Invariance Principle

Validating Causal Abstraction Metrics on Simulated Complex Systems

Coarsening Linear Non-Gaussian Causal Models with Cycles