BriefGPT.xyz
Mar, 2022
无需可微分优化的决策导向学习: 学习局部优化的决策损失
Learning (Local) Surrogate Loss Functions for Predict-Then-Optimize Problems
HTML
PDF
Sanket Shah, Bryan Wilder, Andrew Perrault, Milind Tambe
TL;DR
本文介绍了一种全新的决策化学习方法,通过学习任务特定的损失函数代替了传统的基于代理的优化方法,与先前的工作相比,该方法不需要手工制定基于任务的代理,性能更好且更易用。
Abstract
decision-focused learning
(DFL) is a paradigm for tailoring a
predictive model
to a downstream optimisation task that uses its predictions, so that it can perform better on that specific task. The main technical
→