Metraipolis › archive › dc_68c776397df9
Conformal Bayes under Continuous Label Shift: Sensitivity Analysis and the Limits of Exact Validity — arXiv:2609.27976
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.27976
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T10:58:13.077Z
- SHA-256 of the archived text
72ca9ae515ddf0ba86c079734206c90ef6ca6ed062365353ea8b87349e7442ea- Stable URL of this page
https://metraipolis.com/documento/dc_68c776397df9
How to cite
Conformal Bayes under Continuous Label Shift: Sensitivity Analysis and the Limits of Exact Validity — arXiv:2609.27976 — arXiv (en), https://arxiv.org/abs/2609.27976 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_68c776397df9 (Metraipolis archive dc_68c776397df9, sha256 72ca9ae515ddf0ba86c079734206c90ef6ca6ed062365353ea8b87349e7442ea)
BibTeX
@misc{dc68c776397df9,
title = {Conformal Bayes under Continuous Label Shift: Sensitivity Analysis and the Limits of Exact Validity — arXiv:2609.27976},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.27976},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_68c776397df9; sha256 72ca9ae515ddf0ba86c079734206c90ef6ca6ed062365353ea8b87349e7442ea},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.27976. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Conformal Bayes combines Bayesian posterior predictive scores with conformal calibration, but under continuous label shift both the score and calibration weight depend on the unknown response-marginal density ratio. Existing methods typically estimate one shift parameter from pseudo-labels or predictive samples and plug it into calibration. We instead propose Joint Tilt-Sensitivity Conformal Bayes (JTS-CB), which performs sensitivity analysis over a prespecified set of plausible tilts; its split-conformal realization is JTS-SCB. Each tilt jointly determines the Bayesian conformal score and conformal importance weight. JTS-SCB forms a bounded sensitivity envelope over candidate tilts, but its calibration-only construction does not inherit the exact finite-sample weighted-conformal guarantee. We therefore study a separate candidate-weighted exact counterpart and show that its usefulness depends sharply on tail behavior. For scalar linear exponential tilts, any nonzero candidate tilt makes the exact set unbounded. More generally, tail-growing density ratios produce the same pathology, whereas quadratic tilts with a negative coefficient on \(y^2\) have vanishing tail weights and admit bounded exact inference on the original target. Ratio clipping provides a complementary bounded exact construction for a surrogate target when tails grow. Experiments show that strong plug-in predictive sampling can match the oracle when the shift is well identified, while sensitivity analysis is most useful for richer, weakly identified, or systematically biased shift models, at the cost of wider prediction sets. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-24 in cs.LG). Categorie arXiv: Machine Learning (stat.ML) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.27976 — 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 2026-10-08T10:06:16.832Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-24?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).