AI Paper Atlas

Research map / Probabilistic and scientific modeling / Treatment effects

Local Covariate Selection for Average Causal Effect Estimation without Pretreatment and Causal Sufficiency Assumptions

Zeyu Liu, Zheng Li, Feng Xie, Yan Zeng, Hao Zhang, Kun Zhang

ICML 2026 · Probabilistic and scientific modeling › Treatment effects · approach: Latent-variable causal inference · 0 citations (OpenAlex)

Read the paper

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
use LCS local covariate selection instead of global EHS searchLCS · Jobsrelative bias (%) and nTests: 5.04 -> 3.55 bias; 1732 -> 303 testsImproved
relax pretreatment assumption via LCS instead of LDPLCS · Jobsrelative bias (%): 1635.98 -> 3.55Improved
LCS vs pretreatment-assuming local methods LSAS and CEELSLCS · Jobsrelative bias (%): 11.67 -> 3.55 (LSAS); 22.87 -> 3.55 (CEELS)Improved
LCS needs more CI tests than LSAS, CEELS, LDPLCS · JobsnTests: 176 -> 303 (vs LSAS)Hurt
LCS vs baselines on random graphs (20-50 nodes)LCS · Random graphs (linear Gaussian)relative errorImproved
LCS vs baselines on benchmark networks incl. win95ptsLCS · bnlearn benchmark networksrelative errorImproved
LCS on large networks needs more tests due to initial structure learningLCS · win95ptsnTestsHurt
LCS vs baselines when a treatment descendant is addedLCS · Cattaneo2 (24 variables)estimated effect vs 200-250 g referenceImproved

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