AI Paper Atlas

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

Exploiting Tradeoffs for Exact Recovery in Heterogeneous Stochastic Block Models

Amin Jalali, Qiyang Han, Ioana Dumitriu, Maryam Fazel

NeurIPS 2016 · Learning theory and algorithms › Matrix and tensor methods · approach: High-dimensional estimation theory · 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 per-community relative density rho_k instead of (p_min, n_min) as the summary statisticTheorem 3 (modified ML estimator) · Example 1: two communities n-sqrt(n) and sqrt(n)exact recovery guaranteeImproved
use Theorem 1 conditions with a few very dense tiny communities of size sqrt(log n)SDP convex recovery (2.4) · Example 3: smallest community sizeexact recovery guaranteeImproved
switch from Theorem 1 to Theorem 2 conditions to allow many small communities (smallest size O(log n))SDP convex recovery (2.4) · Example 5: many log n-size communities, spread sizesexact recovery guaranteeMixed
apply Theorem 3 (modified ML) instead of convex Theorems 1 or 2 when p_min - q is O(1)recoverability analysis · Example 6: dense graph, small p_min - qsmallest recoverable community sizeImproved
use SDP convex recovery instead of the peeling strategy of [3]heterogeneous SBM with one big community and many small dense ones · Example 4exact recovery of all communitiesImproved

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