Research map / Learning theory and algorithms / Combinatorial and search optimization
Resilient Constrained Learning
Ignacio Hounie, Alejandro Ribeiro, Luiz F. O. Chamon
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 resilient relaxation instead of hard constraint | Constrained learning · Invariant MNIST, Rotated (90) | test accuracy (%): 96.90 -> 97.19 | Improved |
| use resilient relaxation instead of hard constraint | Constrained learning · Invariant MNIST, Scaled | test accuracy (%): 97.92 -> 98.20 | Improved |
| use resilient relaxation instead of hard constraint | Constrained learning · Invariant MNIST, Translated | test accuracy (%): 93.74 -> 95.21 | Improved |
| use resilient relaxation instead of hard constraint | Constrained learning · Invariant MNIST, Rotated (180) | test accuracy (%): 94.55 -> 95.38 | Improved |
| use resilient relaxation instead of hard constraint | Constrained learning · Invariant MNIST, Original | test accuracy (%): 98.74 -> 98.86 | Improved |
| compare resilient method with Augerino invariance learning | Resilient · Invariant MNIST, Rotated (180) | test accuracy (%): 97.78 -> 95.38 | Hurt |
| compare resilient method with Augerino invariance learning | Resilient · Invariant MNIST, Scaled | test accuracy (%): 97.53 -> 98.20 | No clear effect |
| use resilient relaxation vs constrained baseline in federated learning | Constrained FL · CIFAR10 imbalance 10 | clients improved (%): 77 | Mixed |
| use resilient relaxation vs constrained baseline in federated learning | Constrained FL · F-MNIST imbalance 10 | max deterioration: 23.3 | Hurt |
| increase relaxation cost coefficient alpha (ablation) | Resilient FL · Federated learning, class imbalance | final training loss and perturbation norm | Mixed |
| large alpha makes resilient problem approach constrained problem (ablation) | Resilient FL · Federated learning, class imbalance | perturbation norm | Improved |
| use resilient instead of hard constraint on constraint feasibility | Constrained FL · Federated learning, class imbalance | fraction infeasible clients | Improved |
Closest papers by meaning
- Near-Optimal Solutions of Constrained Learning Problems (ICLR 2024 · similarity 0.87)
- Probably Approximately Correct Constrained Learning (NeurIPS 2020 · similarity 0.86)
- When to retrain a machine learning model (ICML 2025 · similarity 0.83)
- Constrained Diffusion Models via Dual Training (NeurIPS 2024 · similarity 0.83)
- Gradual Domain Adaptation via Manifold-Constrained Distributionally Robust Optimization (NeurIPS 2024 · similarity 0.83)
- Foundation Model-oriented Robustness: Robust Image Model Evaluation with Pretrained Models (ICLR 2024 · similarity 0.83)
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.
