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 1909.06677cs.LGcs.CYstat.ML
收藏

分类中的预测多重性

Predictive Multiplicity in Classification

Haverford College(哈弗福德学院)

Charles T. Marx, Flavio du Pin Calmon, Berk Ustun

摘要:

预测问题通常允许竞争模型表现几乎同样好。当竞争模型分配相互冲突的预测时,这种效应挑战了机器学习中的关键假设。在本文中,我们将预测多重性定义为预测问题接纳具有冲突预测的竞争模型的能力。我们引入正式的措施来评估预测多重性的严重性,并开发整数规划工具来精确计算线性分类问题。我们应用我们的工具来衡量累犯预测问题的预测多重性。我们的研究结果表明,现实世界的数据集可能会承认竞争的模型,分配非常冲突的预测,并激发需要衡量和报告预测的多样性模型的发展。

英文摘要:

Prediction problems often admit competing models that perform almost equally well. This effect challenges key assumptions in machine learning when competing models assign conflicting predictions. In this paper, we define predictive multiplicity as the ability of a prediction problem to admit competing models with conflicting predictions. We introduce formal measures to evaluate the severity of predictive multiplicity and develop integer programming tools to compute them exactly for linear classification problems. We apply our tools to measure predictive multiplicity in recidivism prediction problems. Our results show that real-world datasets may admit competing models that assign wildly conflicting predictions, and motivate the need to measure and report predictive multiplicity in model development.

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。

↑