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.

翻译：预测多样性指分类任务中可能存在多个性能几乎最优但针对个体样本产生冲突输出的竞争模型。这种现象在实际应用中可能引发系统性排斥、无法解释的歧视与不公平等严重问题。然而，由于需要在潜在巨大的假设空间中探索所有此类近乎最优模型（即Rashomon集合），对预测多样性进行度量与缓解面临计算挑战。为解决该问题，我们提出利用Dropout技术探索Rashomon集合中模型的新框架。通过严格的理论推导，建立了Dropout参数与Rashomon集合性质之间的关联，并通过大量实验对框架进行实证评估。数值结果表明，本方法在预测多样性指标估计有效性方面持续优于基线方法，运行速度提升达$20\times \sim 5000\times$。通过高效的Rashomon集合探索与指标估计，进而借助Dropout集成与模型选择实现预测多样性的缓解。