AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Structure Learning of Latent Factors via Clique Search on Correlation Thresholded Graphs

Dale Kim, Qing Zhou

ICML 2023 · Probabilistic and scientific modeling › Molecules and proteins · approach: Sparse graphical and tensor models · 1 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
Replace support learning with the independent maximal clique search (CT) instead of EFA/penalized EFA support learningfactor analysis structure learning · Low-dimensional simulationF1 score of support: near 1.0 vs EFA around 0.5Improved
Use CT clique search rather than penalized EFA for computational efficiencyfactor analysis structure learning · Low-dimensional simulationnumber of models estimated: CT estimates substantially fewer models than EFA-LASSO (30 per d) and EFA-MCP (270 per d)Improved
Violating thresholdability (high-magnitude off-diagonals in Phi) with CT (ablation)CT algorithm high-dimensional simulation · High-dimensional simulation (n=250-1000, p=1.5n)F1 scoreNo clear effect
Violating the unique child condition for CT (ablation)CT algorithm high-dimensional simulation · High-dimensional simulationF1 score and estimated dNo clear effect
Use CT (clique-based structure) on real intelligence data instead of EFAfactor analysis on intelligence test data · Real data (n=301)test-data log-likelihood: CT -3749.60 vs EFA-LASSO -3751.82 and EFA-MCP -3751.37; EFA much worse at -3823.14No 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