Cyclic Counterfactuals under Shift-Scale Interventions

TL;DR

This paper extends counterfactual inference methods to cyclic structural causal models, addressing systems with feedback loops, under shift-scale interventions that modify variable mechanisms through rescaling or shifting.

Contribution

It introduces a framework for counterfactual inference in cyclic SCMs under shift-scale interventions, a setting not well-covered in prior work.

Findings

Develops a theoretical approach for cyclic SCMs

Addresses shift-scale interventions in feedback systems

Provides insights into counterfactual reasoning with cycles

Abstract

Most counterfactual inference frameworks traditionally assume acyclic structural causal models (SCMs), i.e. directed acyclic graphs (DAGs). However, many real-world systems (e.g. biological systems) contain feedback loops or cyclic dependencies that violate acyclicity. In this work, we study counterfactual inference in cyclic SCMs under shift-scale interventions, i.e., soft, policy-style changes that rescale and/or shift a variable's mechanism.