arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2507.23136cs.LG
收藏

观测多重性

Observational Multiplicity

University of California, San Diego(加州大学圣地亚哥分校);University of California, Los Angeles(加州大学洛杉矶分校)

Erin George, Deanna Needell, Berk Ustun

AI总结:

本文提出“观测多重性”概念,通过遗憾度量评估概率分类中模型预测的任意性,并利用弃权与数据收集提升实际应用安全性。

摘要:

许多预测任务可以容纳多个性能几乎相当的模型。当相互竞争的模型对个体给出冲突的预测时,这种现象可能削弱可解释性和安全性。在本工作中,我们研究在概率分类任务中,由于我们称之为“观测多重性”的效应,任意性如何产生。我们讨论了在广泛的实际应用中,当我们学习一个分类器来预测概率 $p_i \in [0,1]$,但给定的是观测值 $y_i \in \{0,1\}$ 的数据集时,这种效应如何出现。我们提出通过“遗憾”的视角来评估个体概率预测的任意性。我们引入了一种用于概率分类任务的遗憾度量,该度量衡量模型的预测可能因不同训练标签而发生的变化。我们提出了一种通用方法来估计概率分类任务中的遗憾。我们使用我们的度量来表明,数据集中某些群体的遗憾通常更高,并讨论了遗憾的潜在应用。我们展示了估计遗憾如何通过弃权(不执行)和数据收集来促进实际应用中的安全性。

英文摘要:

Many prediction tasks can admit multiple models that can perform almost equally well. This phenomenon can undermine interpretability and safety when competing models assign conflicting predictions to individuals. In this work, we study how arbitrariness can arise in probabilistic classification tasks as a result of an effect that we call \emph{observational multiplicity}. We discuss how this effect arises in a broad class of practical applications where we learn a classifier to predict probabilities $p_i \in [0,1]$ but are given a dataset of observations $y_i \in \{0,1\}$. We propose to evaluate the arbitrariness of individual probability predictions through the lens of \emph{regret}. We introduce a measure of regret for probabilistic classification tasks, which measures how the predictions of a model could change as a result of different training labels. We present a general-purpose method to estimate the regret in a probabilistic classification task. We use our measure to show that regret is often higher for certain groups in the dataset and discuss potential applications of regret. We demonstrate how estimating regret can be used to promote safety in real-world applications by abstention and data collection.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑