Conformal prediction beyond exchangeability
Conformal prediction beyond exchangeability
复制标题
DOI：
10.1214/23-aos2276
复制
发表时间：
2022-02
期刊：
影响因子：
--
通讯作者：
R. Barber;E. Candès;Aaditya Ramdas;R. Tibshirani
中科院分区：
文献类型：
--
作者：
R. Barber;E. Candès;Aaditya Ramdas;R. Tibshirani
共形预测是一种流行的现代技术，用于为任意机器学习模型提供有效的预测推理。其有效性依赖于数据可交换性的假设以及给定模型拟合算法作为数据函数的对称性。然而，在实践中部署预测模型时，经常违反交换原则。例如，如果数据分布随着时间的推移而漂移，那么数据点不再是可交换的;此外，在这种情况下，我们可能希望使用非对称算法，将最近的观察视为更相关。本文概括了共形预测来处理这两个方面：我们采用加权分位数来引入对分布漂移的鲁棒性，并设计一种新的随机化技术来允许不对称处理数据点的算法。我们的新方法是可证明的强大，大大减少损失的覆盖面时，交换违反由于分布漂移或其他具有挑战性的功能的真实的数据，同时也实现了相同的覆盖保证现有的共形预测方法，如果数据点实际上是可交换的。我们展示了这些新工具的实际效用与电力和选举预测的模拟和真实数据实验。
Conformal prediction is a popular, modern technique for providing valid predictive inference for arbitrary machine learning models. Its validity relies on the assumptions of exchangeability of the data, and symmetry of the given model fitting algorithm as a function of the data. However, exchangeability is often violated when predictive models are deployed in practice. For example, if the data distribution drifts over time, then the data points are no longer exchangeable; moreover, in such settings, we might want to use a nonsymmetric algorithm that treats recent observations as more relevant. This paper generalizes conformal prediction to deal with both aspects: we employ weighted quantiles to introduce robustness against distribution drift, and design a new randomization technique to allow for algorithms that do not treat data points symmetrically. Our new methods are provably robust, with substantially less loss of coverage when exchangeability is violated due to distribution drift or other challenging features of real data, while also achieving the same coverage guarantees as existing conformal prediction methods if the data points are in fact exchangeable. We demonstrate the practical utility of these new tools with simulations and real-data experiments on electricity and election forecasting.