AI Paper Atlas

Research map / Probabilistic and scientific modeling / Causal discovery

dcFCI: Robust Causal Discovery Under Latent Confounding, Unfaithfulness, and Mixed Data

Adèle H. Ribeiro, Dominik Heider

TPAMI 2026 · Probabilistic and scientific modeling › Causal discovery · approach: Identifiability with hidden confounders · 6 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
increase number of kept compatible PAGs from k=1 to k=2 (ablation)dcFCI · Gaussian, N=10,000True PAG recovery rate (%): 74.7 -> 80.3Improved
increase k from 1 to 2 (ablation)dcFCI · Gaussian, N=50,000True PAG recovery rate (%): 98.3 -> 98.67No clear effect
use data-PAG compatibility score instead of FCI orientationdcFCI vs FCI · Mixed data, N=1,000True PAG recovery rate (%): 5.67 -> 15.67Improved
use dcFCI instead of FCIdcFCI vs FCI · Mixed data, N=5,000True PAG recovery rate (%): 39.67 -> 55Improved
use dcFCI instead of cFCI for validity of output PAGsdcFCI vs cFCI · Gaussian, N=1,000Valid PAGs (%): 46.7 -> 100.0Improved
use dcFCI instead of BCCDdcFCI vs BCCD · Gaussian, N=50,000SHD (mean): 6.7 -> 0.05Improved
use dcFCI instead of FCIdcFCI vs FCI · Gaussian, N=10,000SHD (mean): 1.7 -> 1Improved
use dcFCI instead of MAGSLdcFCI vs MAGSL · Gaussian, N=5,000SHD (mean): 3.4 -> 3.5No clear effect
use dcFCI instead of FCIdcFCI vs FCI · Mixed data, N=1,000SHD (mean): 5.3 -> 4.4Improved
use dcFCI instead of DCDdcFCI vs DCD · Gaussian, N=1,000SHD (mean): 8 -> 5.6Improved
increase number of variables from five to sixdcFCI · Diabetes Health Indicators real dataPAG reliabilityHurt

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