Research map / Learning theory and algorithms / Fairness
FAIRER: Fairness as Decision Rationale Alignment
Tianlin Li, Qing Guo, Aishan Liu, Mengnan Du, Zhiming Li, Yang Liu
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 |
|---|---|---|---|
| add gradient-guided parity alignment (DRAlign) on top of FairReg | Adult MLP · Adult | -DP at similar AP (~0.770): 41.6% DP improvement over second-best (FairReg(ΔDP, Aug)) | Improved |
| increase fairness regularization weight λ from 0 to 0.6 | FairReg(ΔDP, noAug) on Adult · Adult | parity score d_F: 0.624 -> 0.039 | Improved |
| increase fairness regularization weight λ from 0 to 0.6 | FairReg(ΔDP, noAug) on Adult · Adult | AP: accuracy drops by a large margin as λ grows | Hurt |
| add augmentation (FairMixup) to alignment term (ablation) | DRAlign on Credit · Credit | -DP and AP: -0.0169 -> -0.0155; AP 0.877 -> 0.881 | Improved |
| add augmentation to fairness regularization (FairReg Aug) on Credit | FairReg(ΔDP) vs DRAlign on Credit · Credit | -DP: FairReg(ΔDP, Aug) better than DRAlign under DP | Mixed |
| augmentation on CelebA | FairReg(ΔDP) with augmentation · CelebA | -DP: noAug slightly higher fairness than Aug | Hurt |
| oversampling instead of fairness regularization | CelebA wavy hair AlexNet · CelebA | -EO vs AP: oversample -0.069 at AP 0.812 vs FairReg(ΔEO) -0.054 at AP 0.817 | Hurt |
| remove fairness regularization term, keep only alignment loss (w.o.FairReg-DRAlign) (ablation) | Adult MLP · Adult | AP and -DP: (0.776, -0.16) -> (0.781, -0.14) | Improved |
| add alignment loss to fairness regularization on the attractive attribute | CelebA attractive AlexNet · CelebA attractive | -DP and AP: -0.3196 vs -0.3305 at λ=0.1 with AP 0.8956 vs 0.8942 | Improved |
| align only last two layers instead of all layers | ResNet-18 wavy hair · CelebA | -DP: -0.3011 vs -0.3063 at λ=0.1 (DRAlign better AP); -0.2481 vs -0.2552 at λ=10.0 | Mixed |
| increase over-parameterization (hidden size 10 to 200) (ablation) | FairReg(ΔDP, noAug) MLP on Adult · Adult | decision rationale similarity score: c200 reaches only 1.8 at λ=0.6 vs near 3.0 for smaller nets | Hurt |
| add gradient-guided parity alignment on COMPAS (EO metric) | MLP on COMPAS · COMPAS | -EO: -0.044 (DRAlign) vs -0.045 (FairReg ΔEO) at λ=0.05/0.02 comparable AP | No clear effect |
Closest papers by meaning
- Some Optimizers are More Equal: Understanding the Role of Optimizers in Group Fairness (NeurIPS 2025 · similarity 0.89)
- Fair Bilevel Neural Network (FairBiNN): On Balancing fairness and accuracy via Stackelberg Equilibrium (NeurIPS 2024 · similarity 0.87)
- FairSIN: Achieving Fairness in Graph Neural Networks through Sensitive Information Neutralization (AAAI 2024 · similarity 0.87)
- Chasing Fairness in Graphs: A GNN Architecture Perspective (AAAI 2024 · similarity 0.86)
- On the Fairness ROAD: Robust Optimization for Adversarial Debiasing (ICLR 2024 · similarity 0.85)
- On the Alignment between Fairness and Accuracy: from the Perspective of Adversarial Robustness (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.
