Research map / Learning theory and algorithms / Fairness
H-nobs: Achieving Certified Fairness and Robustness in Distributed Learning on Heterogeneous Datasets
Guanqiang Zhou, Ping Xu, Yue Wang, Zhi Tian
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 |
|---|---|---|---|
| increase q-FFL q from 0 to 1 (ablation) | H-nobs, beta=0 · Spambase, no attack | variance of local accuracy: 480 -> 269 | Improved |
| increase q-FFL q from 0 to 1 (ablation) | H-nobs, beta=0 · Spambase, no attack | test accuracy: 92.5% -> 91.3% | Hurt |
| increase q to 2 (ablation) | H-nobs, beta=0 · Spambase, no attack | test accuracy: 91.3% -> 86.9% | Hurt |
| increase screening beta from 0 to 30% at q=0 (ablation) | H-nobs · Spambase, no attack | variance of local accuracy: 480 -> 392 | Improved |
| increase screening beta from 0 to 20% at q=1 (ablation) | H-nobs · Spambase, no attack | variance of local accuracy: 269 -> 331 | Hurt |
| increase screening beta from 0 to 30% (ablation) | H-nobs, q=0 · Spambase, no attack | test accuracy: 92.5% -> 92.2% | No clear effect |
| increase screening beta from 0 to 20% at q=2 (ablation) | H-nobs · Spambase, no attack | variance of local accuracy: 156 -> 243 | Hurt |
| add screening beta=10% at q=2 (ablation) | H-nobs · Law School, no attack | test accuracy: 66.2% -> 73.3% | Improved |
| increase screening beta from 0 to 10% at q=1 (ablation) | H-nobs · Law School, no attack | variance of local accuracy: 353 -> 449 | Hurt |
| increase screening beta from 0 to 30% at q=0 (ablation) | H-nobs · Law School, no attack | variance of local accuracy: 559 -> 507 | Improved |
| increase screening beta from 0 to 30% at q=2 (ablation) | H-nobs · Credit Card Client, no attack | variance of local accuracy: 869 -> 966 | Hurt |
| increase screening beta from 0 to 30% at q=0 (ablation) | H-nobs · Credit Card Client, no attack | variance of local accuracy: 1195 -> 1148 | Improved |
Closest papers by meaning
- Optimal Fair Learning Robust to Adversarial Distribution Shift (ICML 2025 · similarity 0.87)
- Multigroup Robustness (ICML 2024 · similarity 0.87)
- Fairness without Harm: An Influence-Guided Active Sampling Approach (NeurIPS 2024 · similarity 0.87)
- Monotone Individual Fairness (ICML 2024 · similarity 0.86)
- On the Tension between Byzantine Robustness and No-Attack Accuracy in Distributed Learning (ICML 2025 · similarity 0.86)
- Enhancing Robust Fairness via Confusional Spectral Regularization (ICLR 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.
