Current browse context:
cs.CV
Change to browse by:
References & Citations
Computer Science > Computer Vision and Pattern Recognition
Title: Uncertainty Sets for Image Classifiers using Conformal Prediction
(Submitted on 29 Sep 2020 (v1), last revised 3 Sep 2022 (this version, v5))
Abstract: Convolutional image classifiers can achieve high predictive accuracy, but quantifying their uncertainty remains an unresolved challenge, hindering their deployment in consequential settings. Existing uncertainty quantification techniques, such as Platt scaling, attempt to calibrate the network's probability estimates, but they do not have formal guarantees. We present an algorithm that modifies any classifier to output a predictive set containing the true label with a user-specified probability, such as 90%. The algorithm is simple and fast like Platt scaling, but provides a formal finite-sample coverage guarantee for every model and dataset. Our method modifies an existing conformal prediction algorithm to give more stable predictive sets by regularizing the small scores of unlikely classes after Platt scaling. In experiments on both Imagenet and Imagenet-V2 with ResNet-152 and other classifiers, our scheme outperforms existing approaches, achieving coverage with sets that are often factors of 5 to 10 smaller than a stand-alone Platt scaling baseline.
Submission history
From: Anastasios Angelopoulos [view email][v1] Tue, 29 Sep 2020 17:58:04 GMT (1642kb,D)
[v2] Wed, 23 Dec 2020 18:59:13 GMT (1686kb,D)
[v3] Tue, 16 Mar 2021 01:50:26 GMT (3363kb,D)
[v4] Mon, 25 Oct 2021 04:36:04 GMT (3363kb,D)
[v5] Sat, 3 Sep 2022 05:45:19 GMT (2868kb,D)
Link back to: arXiv, form interface, contact.