AI Paper Atlas

Research map / Probabilistic and scientific modeling / Causal discovery

From Temporal to Contemporaneous Iterative Causal Discovery in the Presence of Latent Confounders

Raanan Yehezkel Rohekar, Shami Nisimov, Yaniv Gurwicz, Gal Novik

ICML 2023 · Probabilistic and scientific modeling › Causal discovery · approach: Constraint-based structure learning with latents · 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
Refine temporal relations before contemporaneous relations, testing longer lags first (ablation)TS-ICD · Binary time-seriesCausal accuracy: 0.2158 -> 0.2321; 0.2080 -> 0.2321Improved
Refine temporal relations before contemporaneous relations, testing longer lags first (ablation)TS-ICD · Binary time-seriesCI tests: 388 -> 371; 413 -> 371Improved
Change the CI-test ordering to random (ablation)TS-ICD · Binary time-seriesSkeleton F1: 0.4744 -> 0.4744No clear effect
Change the CI-test ordering to swapped, with contemporaneous edges first (ablation)TS-ICD · Binary time-seriesSkeleton F1: 0.4744 -> 0.4744No clear effect
Use TS-ICD's temporal-first ordering rather than competing algorithms' orderingsCausal discovery algorithms · Binary time seriesSkeleton F1: 0.3602 -> 0.4744; 0.2740 -> 0.4744Improved
Use TS-ICD's temporal-first ordering rather than competing algorithms' orderingsCausal discovery algorithms · Binary time seriesCI tests: 554 -> 371; 550 -> 371; 1351 -> 371Improved
Use TS-ICD on larger graphs instead of LPCMCICausal discovery algorithms · 90-node linear-Gaussian graphsCausal accuracy: 0.16 -> 0.21Improved
Use TS-ICD on larger graphs instead of LPCMCICausal discovery algorithms · 90-node linear-Gaussian graphsCI tests: 11164 -> 8704Improved
Use TS-ICD instead of LPCMCICausal discovery algorithms · Linear-Gaussian time seriesCausal accuracy: 0.2948 -> 0.2500Hurt
Use TS-ICD instead of LPCMCICausal discovery algorithms · Linear-Gaussian time seriesCI tests: 1192 -> 875Improved
Use TS-ICD instead of SVAR-FCI or LPCMCICausal discovery algorithms · Binary time seriesCausal accuracy: 0.2014 -> 0.2321; 0.1443 -> 0.2321Improved

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