Research map / Learning theory and algorithms / Clustering
Guarantees for Spectral Clustering with Fairness Constraints
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Add proportional-representation constraints to spectral clustering | Normalized and unnormalized spectral clustering · FriendshipNet and FacebookNet | Balance: 10% / 15% / 34% / 10% | Helped |
| Add proportional-representation constraints to spectral clustering | Normalized and unnormalized spectral clustering · DrugNet | Balance: 5% / 18% / 86% / 167% | Helped |
| Add proportional-representation constraints to spectral clustering | Normalized and unnormalized spectral clustering · FriendshipNet and FacebookNet | RatioCut or NCut | No clear effect |
| Add proportional-representation constraints to spectral clustering | Normalized and unnormalized spectral clustering · DrugNet (three ethnic groups) | RatioCut or NCut | Hurt |
| Add proportional-representation constraints | Normalized and unnormalized spectral clustering · Synthetic SBM with balanced ground-truth clusters and groups | Error | Helped |
| Use normalized rather than unnormalized spectral clustering with fairness constraints | Fair spectral clustering · Synthetic SBM with balanced ground-truth clusters and groups | Error | Helped |
| Use normalized rather than unnormalized spectral clustering with fairness constraints | Fair spectral clustering · Synthetic SBM with cluster sizes differing or assumptions strongly violated | Error | Helped |
| Increase the number of clusters k | Fair normalized and unnormalized spectral clustering · Synthetic SBM | Error | Hurt |
| Perturb group assignments | Fair normalized and unnormalized spectral clustering · Synthetic SBM, perturbation parameter p | Error: p=0.15; p≥0.7 | Mixed |
Closest papers by meaning
- Accelerating Spectral Clustering under Fairness Constraints (ICML 2025 · similarity 0.93)
- A Differentially Private Clustering Algorithm for Well-Clustered Graphs (ICLR 2024 · similarity 0.89)
- On the Robustness of Spectral Algorithms for Semirandom Stochastic Block Models (NeurIPS 2024 · similarity 0.88)
- Optimal Graph Clustering without Edge Density Signals (NeurIPS 2025 · similarity 0.88)
- Fair Clustering via Alignment (ICML 2025 · similarity 0.87)
- Parameterized Approximation Schemes for Fair-Range Clustering (NeurIPS 2024 · similarity 0.87)
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.