Research map / Learning theory and algorithms / Kernels and regression theory
Hard Shape-Constrained Kernel Machines
Pierre-Cyril Aubin-Frankowski, Zoltan Szabo
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 hard SOC non-crossing constraints instead of PDCD's soft non-crossing regularizer | Joint quantile regression · engel | 100× pinball loss: 48 ± 8 -> 53 ± 9 | Hurt |
| Use hard SOC non-crossing constraints instead of PDCD's soft non-crossing regularizer | Joint quantile regression · GAGurine | 100× pinball loss: 61 ± 7 -> 65 ± 6 | Hurt |
| Use hard SOC non-crossing constraints instead of PDCD's soft non-crossing regularizer | Joint quantile regression · geyser | 100× pinball loss: 105 ± 7 -> 108 ± 3 | Hurt |
| Use hard SOC non-crossing constraints instead of PDCD's soft non-crossing regularizer | Joint quantile regression · mcycle | 100× pinball loss: 66 ± 9 -> 62 ± 5 | Improved |
| Use hard SOC non-crossing constraints instead of PDCD's soft non-crossing regularizer | Joint quantile regression · ftcollinssnow | 100× pinball loss: 154 ± 16 -> 148 ± 13 | Improved |
| Use hard SOC non-crossing constraints instead of PDCD's soft non-crossing regularizer | Joint quantile regression · CobarOre | 100× pinball loss: 159 ± 24 -> 151 ± 17 | Improved |
| Use hard SOC non-crossing constraints instead of PDCD's soft non-crossing regularizer | Joint quantile regression · topo | 100× pinball loss: 69 ± 18 -> 62 ± 14 | Improved |
| Use hard SOC non-crossing constraints instead of PDCD's soft non-crossing regularizer | Joint quantile regression · caution | 100× pinball loss: 88 ± 17 -> 98 ± 22 | Hurt |
| Use hard SOC non-crossing constraints instead of PDCD's soft non-crossing regularizer | Joint quantile regression · ufc | 100× pinball loss: 81 ± 4 -> 87 ± 6 | Hurt |
Closest papers by meaning
- Handling Hard Affine SDP Shape Constraints in RKHSs (JMLR 2022 · similarity 0.91)
- Near-Optimal Solutions of Constrained Learning Problems (ICLR 2024 · similarity 0.84)
- Learning with Exact Invariances in Polynomial Time (ICML 2025 · similarity 0.84)
- Kernel conditional tests from learning-theoretic bounds (NeurIPS 2025 · similarity 0.84)
- The optimality of kernel classifiers in Sobolev space (ICLR 2024 · similarity 0.84)
- Inverse M-Kernels for Linear Universal Approximators of Non-Negative Functions (NeurIPS 2024 · 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.
