BriefGPT.xyz
Feb, 2021
走出“回音壁”：优化表现风险
Outside the Echo Chamber: Optimizing the Performative Risk
HTML
PDF
John Miller, Juan C. Perdomo, Tijana Zrnic
TL;DR
本文探讨了 performative prediction 的问题，基于特定的损失函数和模型诱导的分布位移性质进行优化，从而实现风险凸优化，并且比通用导数优化方法具有更好的样本效率。
Abstract
In
performative prediction
, predictions guide decision-making and hence can influence the distribution of future data. To date, work on
performative prediction
has focused on finding performatively stable models,
→