Research map / Learning theory and algorithms / Prediction and decision losses
PenDer: Incorporating Shape Constraints via Penalized Derivatives
Akhil Gupta, Lavanya Marla, Ruoyu Sun, Naman Shukla, Arinbjörn Kolbeinsson
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 |
|---|---|---|---|
| Enforce increasing monotonicity with LSAT score using derivative penalties | PenDer · Law School Admissions | Monotonicity score: 0.44 ± 0.07 -> 1 | Improved |
| Enforce increasing monotonicity with LSAT score using derivative penalties | PenDer · Law School Admissions | Test accuracy (%): 75.24 ± 0.13 -> 74.87 ± 0.13 | Hurt |
| Use derivative penalties rather than the Lattice model to enforce increasing monotonicity | Shape-constrained models · Law School Admissions | Lipschitz metric (LSAT): 0 -> 0.07 | Improved |
| Enforce decreasing and convex mileage behavior using derivative penalties | PenDer · Used Cars | Test MAE ($): 2, 080 ± 30 -> 1, 982 ± 24 | Improved |
| Enforce decreasing and convex mileage behavior using derivative penalties | PenDer · Used Cars | Monotonicity score: 0.29 ± 0.04 -> 1 | Improved |
| Enforce decreasing and convex mileage behavior using derivative penalties | PenDer · Used Cars | Convexity score: 0.005 ± 0.005 -> 1 | Improved |
| Use derivative penalties rather than the Lattice model to enforce mileage shape | Shape-constrained models · Used Cars | Lipschitz metric: 0 -> 0.012 | Improved |
| Enforce decreasing and convex price behavior using derivative penalties | PenDer · Airline Ancillary | Test accuracy (%): 64.28 ± 0.28 -> 66.39 ± 0.33 | Improved |
| Enforce decreasing and convex price behavior using derivative penalties | PenDer · Airline Ancillary | Monotonicity score: 0.51 ± 0.13 -> 1 | Improved |
| Enforce decreasing and convex price behavior using derivative penalties | PenDer · Airline Ancillary | Convexity score: 0.16 ± 0.04 -> 0.98 ± 0.02 | Improved |
| Use derivative penalties rather than the Lattice model to enforce ancillary-price shape | Shape-constrained models · Airline Ancillary | Lipschitz metric: 0 -> 0.010 ± 0.003 | Improved |
Closest papers by meaning
- Multidimensional Shape Constraints (ICML 2020 · similarity 0.86)
- LCDB 1.1: A Database Illustrating Learning Curves Are More Ill-Behaved Than Previously Thought (NeurIPS 2025 · similarity 0.82)
- Literati: Towards Anytime Optimal Shape Generalized Trees via AO* (NeurIPS 2026 · similarity 0.82)
- Dimension-Free Adaptive Subgradient Methods with Frequent Directions (ICML 2025 · similarity 0.82)
- Empowering Decision Trees via Shape Function Branching (NeurIPS 2025 · similarity 0.82)
- Near-Optimal Solutions of Constrained Learning Problems (ICLR 2024 · similarity 0.82)
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.
