AI Paper Atlas

Research map / Probabilistic and scientific modeling / Treatment effects

Spatial Deconfounder: Interference-Aware Deconfounding for Spatial Causal Inference

Ayush Khot, Miruna Oprescu, Maresa Schröder, Ai Kagawa, Xihaier Luo

ICML 2026 · Probabilistic and scientific modeling › Treatment effects · approach: Proximal and deep causal effect estimation · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add deconfounder substitute confounder (C-VAE) to spatial+ headspatial+ · Local confounding, PM2.5->m (r_d=1), hidden rho_popdirect-effect bias: 0.05 ± 0.02 -> 0.02 ± 0.01Improved
add C-VAE substitute confounder to unet headunet · Spatial confounding, PM2.5->m (r_d=1), hidden rho_popdirect-effect bias: 0.06 ± 0.01 -> 0.04 ± 0.01Improved
add C-VAE substitute confounder to unet headunet · Spatial confounding, SO4->PM2.5 (r_d=1), hidden OCdirect-effect bias: 0.07 ± 0.02 -> 0.06 ± 0.01Improved
add C-VAE substitute confounder to unet headunet · Spatial confounding, SO4->PM2.5 (r_d=1), hidden OCspillover bias: 0.05 ± 0.02 -> 0.07 ± 0.03Hurt
use deconfounder neighborhood radius R=1 instead of R=2 (ablation)C-VAE-spatial+ · Local confounding, PM2.5->m (r_d=1), hidden rho_popdirect-effect bias: 0.04 ± 0.01 -> 0.02 ± 0.01Improved
use deconfounder radius R=2 instead of R=1 (ablation)C-VAE-unet · Spatial confounding, PM2.5->m (r_d=1), hidden rho_popdirect-effect bias: 0.06 ± 0.02 -> 0.04 ± 0.01Improved
use spatial+ head instead of durbinC-VAE-spatial+ vs durbin · Local confounding, SO4->PM2.5 (r_d=2), hidden NH4direct-effect bias: 0.10 ± 0.00 -> 0.02 ± 0.00Improved
compare C-VAE-spatial+ (R=1) to durbin spillover estimateC-VAE-spatial+ vs durbin · Local confounding, PM2.5->m (r_d=1), hidden q_summerspillover bias: 0.02 ± 0.00 -> 0.06 ± 0.00Improved
sparsity sweep 10% to 70% (ablation)Spatial Deconfounder · stress testsbias / predictive p-valueHurt
radius and latent-dimension sweeps (ablation)Spatial Deconfounder · stress testsbiasNo clear effect
inject localized single-cause unobserved confounders (ablation)Spatial Deconfounder · stress testsbiasHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy