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.The code is available at \url{https://github.com/AlvinWen428/FeatureCP}.

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。尽管传统上人们在输出空间中进行共形预测，但这并非唯一可能性。本文提出特征共形预测，通过利用深度表征学习的归纳偏置，将共形预测的适用范围扩展至语义特征空间。从理论角度看，我们证明了在温和假设下，特征共形预测能显著优于常规共形预测。该方法不仅可与基础共形预测结合，还能适配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分类和Cityscapes图像分割等大规模任务中展示了所提方法的最佳性能。代码已开源至 \url{https://github.com/AlvinWen428/FeatureCP}。