AI Paper Atlas

Research map / Probabilistic and scientific modeling / Causal discovery

Switching Regression Models and Causal Inference in the Presence of Discrete Latent Variables

Rune Christiansen, Jonas Peters

JMLR 2020 · Probabilistic and scientific modeling › Causal discovery · approach: Identifiability with hidden confounders · 10 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
enforce equal error variances across hidden states (sigma_Y1^2 = sigma_Y2^2) (ablation)ICPH · Simulated SCM with binary hidden variableidentification of S*: sigma_Y1^2 = sigma_Y2^2 vs sigma_Y1^2, sigma_Y2^2 >= 10^{-4}Improved
increase sample sizeICPH · Simulated SCM with binary hidden variableidentification of S*: n from 100 to 500Improved
lower GMEP (harder hidden-state estimation)adjusted confidence regions · Simulated SR model, n=100empirical coverage of 95% confidence regionsHurt
use ICPH instead of k-means ICP and JCI-PCcausal discovery · Simulated SCM with binary hidden variablelevel and powerImproved
initialize optimizer at the true parameter values instead of data-driven (ablation)NLM optimizer · Simulated SR model, n=100likelihood score: less than 0.2%No clear effect
use EM instead of NLM optimizer (ablation)ICPH · Simulated dataresults and runtime: computation time larger by approximately a factor of 6No clear effect

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