Conformal prediction is a distribution-free technique for establishing valid prediction intervals. Although conventionally people conduct conformal prediction in the output space, this is not the only possibility. In this paper, we propose feature conformal prediction, which extends the scope of conformal prediction to semantic feature spaces by leveraging the inductive bias of deep representation learning. From a theoretical perspective, we demonstrate that feature conformal prediction provably outperforms regular conformal prediction under mild assumptions. Our approach could be combined with not only vanilla conformal prediction, but also other adaptive conformal prediction methods. Apart from experiments on existing predictive inference benchmarks, we also demonstrate the state-of-the-art performance of the proposed methods on large-scale tasks such as ImageNet classification and Cityscapes image segmentation.

翻译：共形预测是一种无需分布假设即可构建有效预测区间的技术。尽管传统方法在输出空间中进行共形预测，但这并非唯一途径。本文提出特征共形预测，通过利用深度表示学习的归纳偏置，将共形预测的应用范围扩展至语义特征空间。从理论角度，我们证明在温和假设条件下，特征共形预测在性能上严格优于常规共形预测。该方法不仅能与原始共形预测结合，还可兼容其他自适应共形预测方法。除在现有预测推断基准上的实验外，我们还在ImageNet图像分类、Cityscapes图像分割等大规模任务中展示了所提方法的最新性能成果。