AI Paper Atlas

Research map / Graph learning / Graph algorithms and clustering

Cluster Randomized Designs for One-Sided Bipartite Experiments

Jennifer Rogers Brennan, Vahab Mirrokni, Jean Pouget-Abadie

NeurIPS 2022 · Graph learning › Graph algorithms and clustering · approach: Neural and local graph clustering · 6 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
use bias-minimizing clustering H(C) instead of direct bipartite clustering (ablation)cluster-randomized design · Bipartite SBM p=0.005relative bias: 9.10 -> 3.90Improved
use H(C) clustering instead of unit-level randomizationcluster-randomized design · Bipartite SBM p=0.0relative bias: 12.44 -> 0.02Improved
H(C) vs true clusteringcluster-randomized design · Bipartite SBMrelative bias: 3.88 -> 3.90No clear effect
replace H(C) objective with Tr(Var(d)) objective (ablation)cluster-randomized design · Bipartite SBMrelative bias: 3.90 -> 3.84No clear effect
increase cross-cluster edge probability p from 0 to 0.5H(C) design · Bipartite SBMrelative bias: 0.02 -> 12.98Hurt
widen pure-exposure neighborhood Delta from 0.1 to 0.5H(C) design · Nonlinear outcomesrelative bias: 1.000 -> 0.001Improved
use H(C) instead of direct clustering at Delta=0.5 (ablation)cluster-randomized design · Nonlinear outcomesrelative bias: 0.600 -> 0.001Improved
use H(C) clustering instead of EXPOSURE-DESIGN at Delta=0.3 (ablation)cluster-randomized design · Nonlinear outcomesrelative bias: 0.509 -> 0.457Improved
use H(C) instead of unit-level randomization on power-law graphs (strong latent structure)cluster-randomized design · Power-law graphrelative bias: 4.9 -> 1.9Improved
use H(C) instead of true clusters on power-law graphs (weak structure) (ablation)cluster-randomized design · Power-law graphrelative bias: 5.2 -> 3.5Improved
use H(C) vs direct clustering on power-law graphs (ablation)cluster-randomized design · Power-law graphrelative bias: 3.3 -> 3.5No clear effect
use H(C) instead of direct clustering in Airbnb misspecified-exposure simulation (ablation)cluster-randomized design · Airbnb marketplace simulationRMSEImproved

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