AI Paper Atlas

Research map / Graph learning / Graph algorithms and clustering

Learning to Induce Causal Structure

Nan Rosemary Ke, Silvia Chiappa, Jane X Wang, Jorg Bornschein, Anirudh Goyal, Melanie Rey, Theophane Weber, Matthew Botvinick, Michael Curtis Mozer, Danilo Jimenez Rezende

ICLR 2023 · Graph learning › Graph algorithms and clustering · approach: Continuous DAG structure learning · 17 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
train CSIvA on synthetic Dirichlet data and apply it to SachsCSIvA · SachsHamming distance: 4 vs 13Improved
train CSIvA on synthetic MLP data and apply it to AsiaCSIvA · AsiaHamming distance: 3 vs 4Improved
use CSIvA rather than Non-linear ICP on nonlinear MLP graphsCausal structure induction · MLP synthetic, ER-2, N=10Hamming distance: 5.12 vs 9.12Improved
use CSIvA rather than Non-linear ICP on continuous graphsCausal structure induction · Continuous synthetic, ER-2, N=10Hamming distance: 1.73 vs 2.41Improved
use CSIvA rather than DAG-GNN on Dirichlet graphsCausal structure induction · Dirichlet synthetic, ER-2, N=10, α=1.0Hamming distance: 4.59 vs 9.3Improved
train on Dirichlet data with α=5.0 rather than α≤0.5 (ablation)CSIvA · Dirichlet synthetic, N=10Hamming distance: 4.91 vs 0.72Hurt
use 1500 rather than 1000 samples per CBN (ablation)CSIvA · Dirichlet synthetic, ER-2 graphsHamming distanceImproved
use 1000 samples per CBN rather than fewer samples (ablation)CSIvA · Dirichlet synthetic, ER-1 graphsHamming distanceImproved
train on ER-2 rather than ER-1 graphs (ablation)CSIvA on MLP data · MLP synthetic, N=7Hamming distance: ER-1: 1.2 -> 3.3; ER-2: 1.8 -> 1.8; ER-3: 5.0 -> 2.8Mixed
train with Dirichlet α=0.25 rather than α=0.1 (ablation)CSIvA · Dirichlet synthetic, ER-1, N=7Hamming distance: 0.31 -> 0.33; 0.72 -> 0.40; 1.8 -> 0.71Mixed
increase Dirichlet α from 0.1–0.5 to 1.0 or 5.0 (ablation)CSIvA · Dirichlet syntheticHamming distance: 0.72 -> 2.31; 1.27 -> 4.91Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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