Diffusion Causal Models for Counterfactual Estimation

2022/02/21 by Pedro L. Sánchez, Sotirios A. Tsaftaris, Sanchez, Pedro +1 · 12 citations
Computer Science · #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Generative Adversarial Networks and Image Synthesis #Machine Learning (cs.LG) #Machine Learning in Healthcare

paper · pdf · doi:10.48550/arxiv.2202.10166

openalex publication_date 2022/02/21 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

We consider the task of counterfactual estimation from observational imaging data given a known causal structure. In particular, quantifying the causal effect of interventions for high-dimensional data with neural networks remains an open challenge. Herein we propose Diff-SCM, a deep structural causal model that builds on recent advances of generative energy-based models. In our setting, inference is performed by iteratively sampling gradients of the marginal and conditional distributions entailed by the causal model. Counterfactual estimation is achieved by firstly inferring latent variables with deterministic forward diffusion, then intervening on a reverse diffusion process using the gradients of an anti-causal predictor w.r.t the input. Furthermore, we propose a metric for evaluating the generated counterfactuals. We find that Diff-SCM produces more realistic and minimal counterfactuals than baselines on MNIST data and can also be applied to ImageNet data. Code is available https://github.com/vios-s/Diff-SCM.

Cited by

• The Causal Round Trip: Generating Authentic Counterfactuals by Eliminating Information Loss
• Cyclic Counterfactuals under Shift-Scale Interventions
• Towards Error-Centric Intelligence II: Energy-Structured Causal Models
• Spatial Deconfounder: Interference-Aware Deconfounding for Spatial Causal Inference
• Counterfactual Identifiability via Dynamic Optimal Transport
• Causal-Adapter: Taming Text-to-Image Diffusion for Faithful Counterfactual Generation
• GDR-learners: Orthogonal Learning of Generative Models for Potential Outcomes
• V-CECE: Visual Counterfactual Explanations via Conceptual Edits
• DoubleGen: Debiased Generative Modeling of Counterfactuals
• Counterfactual Probabilistic Diffusion with Expert Models
• A Selective Review of Modern Stochastic Modeling: SDE/SPDE Numerics, Data-Driven Identification, and Generative Methods with Applications in Biomathematics
• Annotation-Assisted Learning of Treatment Policies From Multimodal Electronic Health Records

Related

• Causal Diffusion Autoencoders: Toward Counterfactual Generation via Diffusion Probabilistic Models
• DiffusionCounterfactuals: Inferring High-dimensional Counterfactuals with Guidance of Causal Representations
• Deep Structural Causal Models for Tractable Counterfactual Inference
• High Fidelity Image Counterfactuals with Probabilistic Causal Models
• Counterfactual Probabilistic Diffusion with Expert Models