Research map / Learning theory and algorithms / Combinatorial and search optimization
Certified Monotonic Neural Networks
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 |
|---|---|---|---|
| train end-to-end instead of freezing the pretrained feature extractor (ablation) | Certified monotonic network · Chest X-Ray | test accuracy: 62.3% ± 0.2% -> 66.3% ± 1.0% | Helped |
| certified monotonic regularized network vs Non-Neg-DNN with same structure | two-layer ReLU nets · COMPAS | test accuracy: 67.3% ± 0.9% -> 68.8% ± 0.2% | Helped |
| certified monotonic regularized network vs Non-Neg-DNN with same structure | ReLU nets · Blog Feedback | RMSE: 0.168 ± 0.001 -> 0.158 ± 0.001 | Helped |
| certified monotonic regularized network vs Non-Neg-DNN with same structure | ReLU nets · Loan Defaulter | test accuracy: 65.1% ± 0.1% -> 65.2% ± 0.1% | No clear effect |
| certified monotonic regularized network vs DLN | monotonic nets · Chest X-Ray | test accuracy: 65.4% ± 0.1% -> 66.3% ± 1.0% | Helped |
| add monotonicity constraints to CNN trained on MNIST binary tasks (ablation) | CNN · MNIST pairs under PGD attack | adversarial test accuracy | Helped |
| add monotonicity constraints to CNN trained on MNIST binary tasks (ablation) | CNN · MNIST pairs, clean images | test accuracy | No clear effect |
Closest papers by meaning
- Deep Lattice Networks and Partial Monotonic Functions (NeurIPS 2017 · similarity 0.90)
- Scalable Monotonic Neural Networks (ICLR 2024 · similarity 0.90)
- Smooth Min-Max Monotonic Networks (ICML 2024 · similarity 0.89)
- Advancing Constrained Monotonic Neural Networks: Achieving Universal Approximation Beyond Bounded Activations (ICML 2025 · similarity 0.88)
- Learning Monotonic Probabilities with a Generative Cost Model (ICML 2025 · similarity 0.85)
- Generalization Bounds for Rank-sparse Neural Networks (NeurIPS 2025 · similarity 0.84)
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.