English(EN) Decision-Focused Learning via Tangent-Space Projection of Prediction Error

新方法将预测误差投影到切空间以实现面向决策的学习

作者 PulseAugur 编辑部 · [1 个来源] · 2026-05-05 04:00

研究人员开发了一种名为PEAR（Projected Error As Regret-gradient）的新方法，用于面向决策的学习。该技术通过将预测误差投影到活动约束的切空间上，简化了遗憾梯度的计算。与依赖求解器微分或使用代理损失的现有方法相比，PEAR提供了一种计算效率更高、更直接的替代方案。实验表明，即使约束发生变化，PEAR在优化基准和实际任务上也能获得卓越的决策质量。 AI

影响引入了一种更有效的方法来训练模型，以提高优化任务中下游决策的质量。

排序理由这是一篇介绍面向决策学习新方法的学术论文。

在 arXiv cs.LG 阅读 →

Decision-Focused Learning

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 0 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 这是一篇介绍面向决策学习新方法的学术论文。
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, other; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: 150 days old; Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.LG TIER_1 English(EN) · Junhyeong Lee, Sangjin Jin, Yongjae Lee · 2026-05-05 04:00

通过预测误差的切线空间投影实现决策导向学习

arXiv:2605.01361v1 Announce Type: new 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…

本文如何被排名

报道来源 [1]

通过预测误差的切线空间投影实现决策导向学习

相关实体

相关话题