AI Paper Atlas

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

AAAI 2021 · Learning theory and algorithms › Prediction and decision losses · approach: Theory of neural network sensitivity · 0 citations (OpenAlex)

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 changedWhereMeasuredEffect
Enforce increasing monotonicity with LSAT score using derivative penaltiesPenDer · Law School AdmissionsMonotonicity score: 0.44 ± 0.07 -> 1Improved
Enforce increasing monotonicity with LSAT score using derivative penaltiesPenDer · Law School AdmissionsTest accuracy (%): 75.24 ± 0.13 -> 74.87 ± 0.13Hurt
Use derivative penalties rather than the Lattice model to enforce increasing monotonicityShape-constrained models · Law School AdmissionsLipschitz metric (LSAT): 0 -> 0.07Improved
Enforce decreasing and convex mileage behavior using derivative penaltiesPenDer · Used CarsTest MAE ($): 2, 080 ± 30 -> 1, 982 ± 24Improved
Enforce decreasing and convex mileage behavior using derivative penaltiesPenDer · Used CarsMonotonicity score: 0.29 ± 0.04 -> 1Improved
Enforce decreasing and convex mileage behavior using derivative penaltiesPenDer · Used CarsConvexity score: 0.005 ± 0.005 -> 1Improved
Use derivative penalties rather than the Lattice model to enforce mileage shapeShape-constrained models · Used CarsLipschitz metric: 0 -> 0.012Improved
Enforce decreasing and convex price behavior using derivative penaltiesPenDer · Airline AncillaryTest accuracy (%): 64.28 ± 0.28 -> 66.39 ± 0.33Improved
Enforce decreasing and convex price behavior using derivative penaltiesPenDer · Airline AncillaryMonotonicity score: 0.51 ± 0.13 -> 1Improved
Enforce decreasing and convex price behavior using derivative penaltiesPenDer · Airline AncillaryConvexity score: 0.16 ± 0.04 -> 0.98 ± 0.02Improved
Use derivative penalties rather than the Lattice model to enforce ancillary-price shapeShape-constrained models · Airline AncillaryLipschitz metric: 0 -> 0.010 ± 0.003Improved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy