atlas

Research map / Learning theory and algorithms / Matrix and tensor methods

On Mixed Memberships and Symmetric Nonnegative Matrix Factorizations

ICML 2017 · Learning theory and algorithms › Matrix and tensor methods · approach: Matrix completion and factorization · 10 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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
Use GeoNMF instead of SVIMMSB inference · Simulated MMSB networksrelative parameter errorHelped
Use GeoNMF instead of SVIMMSB inference · Simulated networks with sparse graphsrelative parameter errorHelped
Use GeoNMF instead of SVIMMSB inference · Simulated networkswall-clock time: about 100 times slowerHelped
Use GeoNMF instead of competing methodsOverlapping community detection · Facebook and G-plus networksSpearman rank correlationNo clear effect
Use GeoNMF instead of competing methodsOverlapping community detection · DBLP co-authorship networksSpearman rank correlationMixed
Use GeoNMF instead of competing methodsOverlapping community detection · MAG co-authorship networkswall-clock timeHelped
Increase diagonal interaction skewness (ablation)GeoNMF · Simulated MMSB networksrelative parameter errorNo clear effect
Add noise to off-diagonal entries of B (ablation)GeoNMF · Simulated MMSB networksrelative parameter errorNo clear effect
Increase community overlap via larger alpha_0 (ablation)GeoNMF · Simulated MMSB networksrelative parameter errorHurt
Increase community overlap via larger alpha_0 (ablation)GeoNMF versus SVI · Simulated MMSB networksrelative parameter errorHelped
Decrease graph sparsity parameter rho (ablation)GeoNMF versus SVI · Simulated MMSB networksrelative parameter errorHelped
Use sample splitting and cross-partition projection (ablation)GeoNMF · MMSB parameter recoveryrow-wise estimation errorHelped

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