Predictive Inference with Feature Conformal Prediction
This work addresses the need for more accurate predictive uncertainty estimation in machine learning, particularly for high-stakes applications, though it is incremental as it builds on existing conformal prediction methods.
The paper tackles the problem of improving prediction intervals by extending conformal prediction to feature spaces, demonstrating that feature conformal prediction provably outperforms regular conformal prediction and achieves state-of-the-art performance on large-scale tasks like ImageNet classification and Cityscapes segmentation.
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}.