Research map / Learning theory and algorithms / Prediction and decision losses
Monotonic Calibrated Interpolated Look-Up Tables
Maya Gupta, Andrew Cotter, Jan Pfeifer, Konstantin Voevodski, Kevin Canini, Alexander Mangylov, Wojciech Moczydlowski, Alexander van Esbroeck
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 |
|---|---|---|---|
| randomly sample regularizer terms per SGD step (k_R=1024, k_l=1) (ablation) | torsion-regularized calibrated lattice SGD · real-world application | training time and held-out accuracy: 150x speed-up | Improved |
| use LightTouch stochastic constraints instead of suboptimal projection (ablation) | monotonic lattice SGD · constraint handling | convergence speed and results | No clear effect |
| use suboptimal active-set projection instead of full projection (ablation) | monotonic lattice SGD · constraint handling | objective value and test accuracy | No clear effect |
| add adagrad step sizes (ablation) | projected SGD for monotonic lattices · training convergence | convergence speed | Mixed |
| use simplex instead of multilinear interpolation on 2^D lattices (ablation) | lattice regression · case studies, D >= 8 | accuracy and speed | No clear effect |
Closest papers by meaning
- Fast and Flexible Monotonic Functions with Ensembles of Lattices (NeurIPS 2016 · similarity 0.93)
- Improving Multi-Class Calibration through Normalization-Aware Isotonic Techniques (ICML 2025 · similarity 0.85)
- Smooth Min-Max Monotonic Networks (ICML 2024 · similarity 0.84)
- Advancing Constrained Monotonic Neural Networks: Achieving Universal Approximation Beyond Bounded Activations (ICML 2025 · similarity 0.83)
- Learning Monotonic Probabilities with a Generative Cost Model (ICML 2025 · similarity 0.83)
- Robustly Learning Monotone Single-Index Models (NeurIPS 2025 · 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.
