Abstract:Decision-Focused Learning (DFL) trains predictors to improve downstream decision quality, but computing regret gradients typically requires differentiating through solvers or relying on surrogate losses, which can be computationally expensive or deviate from the true objective. We show that, under standard regularity with locally stable active constraints, the regret gradient admits a closed-form geometric characterization, equivalent to the prediction error projected onto the tangent space of active constraints, scaled by local curvature. This reveals that regret gradients can be obtained by filtering decision-irrelevant components from the MSE gradient, providing a simpler and more direct alternative to existing approaches. Based on this, we propose PEAR (Projected Error As Regret-gradient), which computes regret gradients via a reduced linear system over active constraints, avoiding differentiation through solver iterations or additional optimization solves. Experiments on LP benchmarks and a real-world QP task show that PEAR achieves the best decision quality among all baselines while being the most computationally efficient, with gains that persist under constraint shifts.
| Comments: | 20 pages, 4 figures, 8 tables |
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2605.01361 [cs.LG] |
| (or arXiv:2605.01361v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2605.01361 arXiv-issued DOI via DataCite (pending registration) |
From: Junhyeong Lee [view email]
[v1]
Sat, 2 May 2026 10:10:16 UTC (518 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。