Research map / Learning theory and algorithms / Clustering
Change Point Estimation in a Dynamic Stochastic Block Model
Monika Bhattacharjee, Moulinath Banerjee, George Michailidis
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use 2-step estimator instead of every-time-point clustering | change-point estimation · DSBM m=60,n=60, reallocation delta=1/4 | exact change-point recoveries out of 100: 34 -> 80 | Improved |
| use 2-step estimator instead of every-time-point clustering | change-point estimation · DSBM m=60,n=60, change in connectivity | exact change-point recoveries out of 100: 40 -> 83 | Improved |
| use 2-step estimator instead of every-time-point clustering | change-point estimation · DSBM m=60,n=60, merging communities | exact change-point recoveries out of 100: 21 -> 88 | Improved |
| use 2-step estimator instead of every-time-point clustering | change-point estimation · DSBM m=60,n=60, lambda=1/20, reallocation delta=1/10 | exact change-point recoveries out of 100: 23 -> 87 | Improved |
| use 2-step estimator instead of every-time-point clustering | change-point estimation · DSBM m=60,n=60, lambda=1/20, merging communities | exact change-point recoveries out of 100: 21 -> 88 | Improved |
| use 2-step estimator instead of every-time-point clustering | change-point estimation · DSBM m=500,n=20, reallocation delta=1/20 | exact change-point recoveries out of 100: 85 -> 90 | No clear effect |
| use 2-step estimator instead of known-community estimator | change-point estimation · DSBM m=500,n=20, reallocation delta=1/4 | exact change-point recoveries out of 100: 90 -> 39 | Hurt |
| use 2-step estimator instead of known-community estimator | change-point estimation · DSBM m=60,n=60, reallocation delta=1/20 | exact change-point recoveries out of 100: 90 -> 88 | No clear effect |
| increase nodes m from 60 to 500 (n=20) | every-time-point clustering · DSBM merging communities | exact change-point recoveries out of 100: 21 -> 80 | Improved |
| increase nodes m from 60 to 500 (n=20) | every-time-point clustering · DSBM change in connectivity | exact change-point recoveries out of 100: 40 -> 83 | Improved |
| increase signal strength by reducing delta from 1/4 to 1/20 | every-time-point clustering · DSBM m=60,n=60, reallocation | exact change-point recoveries out of 100: 34 -> 85 | Improved |
Closest papers by meaning
- Autoregressive Networks (JMLR 2023 · similarity 0.90)
- Efficient Near-Optimal Testing of Community Changes in Balanced Stochastic Block Models (NeurIPS 2019 · similarity 0.89)
- Determining the Number of Communities in Degree-corrected Stochastic Block Models (JMLR 2021 · similarity 0.88)
- Bayesian Model Selection for Change Point Detection and Clustering (ICML 2018 · similarity 0.88)
- Multiresolution Analysis and Statistical Thresholding on Dynamic Networks (NeurIPS 2025 · similarity 0.86)
- Revisiting Instance-Optimal Cluster Recovery in the Labeled Stochastic Block Model (ICML 2025 · similarity 0.85)
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.
