Predictive multiplicity refers to the phenomenon in which classification tasks may admit multiple competing models that achieve almost-equally-optimal performance, yet generate conflicting outputs for individual samples. This presents significant concerns, as it can potentially result in systemic exclusion, inexplicable discrimination, and unfairness in practical applications. Measuring and mitigating predictive multiplicity, however, is computationally challenging due to the need to explore all such almost-equally-optimal models, known as the Rashomon set, in potentially huge hypothesis spaces. To address this challenge, we propose a novel framework that utilizes dropout techniques for exploring models in the Rashomon set. We provide rigorous theoretical derivations to connect the dropout parameters to properties of the Rashomon set, and empirically evaluate our framework through extensive experimentation. Numerical results show that our technique consistently outperforms baselines in terms of the effectiveness of predictive multiplicity metric estimation, with runtime speedup up to $20\times \sim 5000\times$. With efficient Rashomon set exploration and metric estimation, mitigation of predictive multiplicity is then achieved through dropout ensemble and model selection.
Be Intentional About Fairness!: Fairness, Size, and Multiplicity in the Rashomon Set
New York University（纽约大学）；NYU Courant Institute of Mathematical Sciences（纽约大学柯朗数学科学研究所）；Robert F. Wagner Graduate School of Public Service, New York University（纽约大学罗伯特·F·瓦格纳公共服务研究生院）；Tandon School of Engineering, New York University（纽约大学坦登工程学院）；Center for Urban Science and Progress, Tandon School of Engineering, New York University（纽约大学坦登工程学院城市科学与进步中心）；Center for Data Science, New York University（纽约大学数据科学中心）
An Experimental Study on the Rashomon Effect of Balancing Methods in Imbalanced Classification
Eskisehir Technical University, Department of Statistics, Turkiye（埃斯基谢普技术大学统计系，土耳其）；Warsaw University of Technology, Faculty of Mathematics and Information Science（华沙技术大学数学与信息科学系）；University of Warsaw, Faculty of Mathematics, Informatics and Mechanics（华沙大学数学、信息学与力学系）
REALITrees: Rashomon Ensemble Active Learning for Interpretable Trees
Department of Statistics, University of Washington（统计学系，华盛顿大学）；Department of Computer Science, Duke University（计算机科学系，杜克大学）；Department of Sociology, University of Washington（社会学系，华盛顿大学）
Horizon-Constrained Rashomon Sets for Chaotic Forecasting
Dept. of Electrical Engineering, California State University Long Beach, USA；WorkOnward Inc., USA；Dept. of Computer Engineering \& Computer Science, California State University Long Beach, USA
Investigating the Impact of Balancing, Filtering, and Complexity on Predictive Multiplicity: A Data-Centric Perspective
Eskisehir Technical University, Department of Statistics（埃斯克谢希大学统计系）；Warsaw University of Technology, Faculty of Mathematics and Information Science（华沙技术大学数学与信息科学学院）；University of Warsaw, Faculty of Mathematics, Informatics and Mechanics（华沙大学数学、信息学与力学学院）
Beyond the Single-Best Model: Rashomon Partial Dependence Profile for Trustworthy Explanations in AutoML
Department of Statistics, Eskisehir Technical University, Turkiye（埃斯基谢普大学统计系）；Leiden Institute of Advanced Computer Science, Leiden University, the Netherlands（莱顿大学高级计算机科学研究所）；Faculty of Mathematics and Information Science, Warsaw University of Technology, Poland（华沙理工大学数学与信息科学学院）；Informatics and Mechanics, University of Warsaw, Faculty of Mathematics, Poland（华沙大学信息技术与力学系）
