Metraipolis › archive › dc_19fa9bf9df9d
CalCErt: Bin-wise Certification of Confidence Calibration in Medical Image Classification — arXiv:2610.03142
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.03142
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T09:07:52.808Z
- SHA-256 of the archived text
eaf485b9c8ff7863b146c389ae754b50bd164ca70bd9a7c87c8078a69a4944b7- Stable URL of this page
https://metraipolis.com/documento/dc_19fa9bf9df9d
How to cite
CalCErt: Bin-wise Certification of Confidence Calibration in Medical Image Classification — arXiv:2610.03142 — arXiv (en), https://arxiv.org/abs/2610.03142 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_19fa9bf9df9d (Metraipolis archive dc_19fa9bf9df9d, sha256 eaf485b9c8ff7863b146c389ae754b50bd164ca70bd9a7c87c8078a69a4944b7)
BibTeX
@misc{dc19fa9bf9df9d,
title = {CalCErt: Bin-wise Certification of Confidence Calibration in Medical Image Classification — arXiv:2610.03142},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.03142},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_19fa9bf9df9d; sha256 eaf485b9c8ff7863b146c389ae754b50bd164ca70bd9a7c87c8078a69a4944b7},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.03142. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Deep neural networks remain vulnerable to adversarial perturbations, which can distort not only predictions but also confidence scores, undermining uncertainty calibration. While existing certification methods focus on preserving the predicted category, providing guarantees on how calibration behaves under adversarial attacks remains overlooked. In this work, we introduce CalCErt, a simple and efficient post-hoc strategy that certifies bin-wise confidence calibration for any pretrained differentiable classifier. Our approach combines empirical calibration estimates, statistical concentration bounds, and local Lipschitz estimates of the confidence function to derive data-dependent upper bounds on worst-case miscalibration within an $ell_2$-ball of radius R. We evaluate CalCErt across 11 medical image classification tasks and multiple adversarial perturbations, demonstrating substantially higher certified coverage than baseline strategies while maintaining competitive tightness. Our code is available at this https URL . Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-05 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2610.03142 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-06T08:21:44.211Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).