Research map / Learning theory and algorithms / Fairness
Learning Fair Representations with Kolmogorov-Arnold Networks
Amisha Priyadarshini, Sergio Gago-Masague
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 ADOPT optimizer instead of Adam for KAN adversarial debiasing (ablation) | KAN adversarial debiasing · UCI admissions D(2) | p%-Rule (low-income): 84.11 -> 99.25 | Improved |
| use OAdam instead of Adam for KAN (ablation) | KAN adversarial debiasing · UCI admissions D(1) | p%-Rule (low-income): 85.79 -> 90.39; AUROC 80.27 -> 76.86 | Mixed |
| use ADOPT instead of Adam for KAN (ablation) | KAN adversarial debiasing · UCI admissions D(1) | AUROC: 80.27 -> 81.45 | Improved |
| train KAN on larger D(2) instead of D(1) | KAN adversarial debiasing · UCI admissions | accuracy: 74.92 -> 82.36 | Improved |
| replace FFNN with KAN in adversarial debiasing | FFNN adversarial debiasing · UCI admissions D(2) | accuracy: 74.87 -> 82.36 | Improved |
| replace FFNN with KAN in adversarial debiasing | FFNN adversarial debiasing · UCI admissions D(1) | p%-Rule (low-income): 98.32 -> 85.79 | Hurt |
| use KAN adversarial debiasing instead of ROAD | ROAD B(3) · UCI admissions D(1) | p%-Rule (low-income): 83.10 -> 90.39 | Improved |
| increase B-spline order k from 3 to 5 (ablation) | KAN adversarial debiasing · UCI admissions D(1) | p%-Rule | Mixed |
| add Lipschitz/smooth spline-based KAN structure to adversary and classifier | KAN adversarial debiasing · UCI admissions D(1) | training loss | Hurt |
Closest papers by meaning
- Adversarial Fairness Network (AAAI 2024 · similarity 0.89)
- Fair Representation Learning with Controllable High Confidence Guarantees via Adversarial Inference (NeurIPS 2025 · similarity 0.89)
- On the Fairness ROAD: Robust Optimization for Adversarial Debiasing (ICLR 2024 · similarity 0.89)
- FairNet: Dynamic Fairness Correction without Performance Loss via Contrastive Conditional LoRA (NeurIPS 2025 · similarity 0.89)
- On the Alignment between Fairness and Accuracy: from the Perspective of Adversarial Robustness (ICML 2025 · similarity 0.88)
- DAFA: Distance-Aware Fair Adversarial Training (ICLR 2024 · similarity 0.88)
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.
