Metraipolis › archive › dc_e21994b3ede0
Augmented Feature Boosting for Multicalibration — arXiv:2609.33891
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.33891
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T22:37:50.711Z
- SHA-256 of the archived text
8bece55efc08d4b976e282e12bb21dc024c6301c1e3339a1ea03be18e7d5c99f- Stable URL of this page
https://metraipolis.com/documento/dc_e21994b3ede0
How to cite
Augmented Feature Boosting for Multicalibration — arXiv:2609.33891 — arXiv (en), https://arxiv.org/abs/2609.33891 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e21994b3ede0 (Metraipolis archive dc_e21994b3ede0, sha256 8bece55efc08d4b976e282e12bb21dc024c6301c1e3339a1ea03be18e7d5c99f)
BibTeX
@misc{dce21994b3ede0,
title = {Augmented Feature Boosting for Multicalibration — arXiv:2609.33891},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.33891},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e21994b3ede0; sha256 8bece55efc08d4b976e282e12bb21dc024c6301c1e3339a1ea03be18e7d5c99f},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33891. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Multicalibration requires a predictor's residuals to be unbiased not only globally, but also after conditioning on the predictor's own level sets and reweighting by a rich class of test functions. Standard boosting approaches in the distributional setting achieve this by repeatedly discretizing the predictor's range then auditing and repairing the resulting level sets. One consequence is that in practice, the algorithm's guarantees are sensitive to this parametrization of the rounding parameter. A natural theoretical question, then, is how to do discretization-free boosting which avoids this rounding within the boosting process itself. Here, we analyze an alternative feature-augmentation boosting paradigm inspired by Tax et al. (2026): at each round, a squared-loss oracle is called on hypotheses that receive the previous predictor's output as an additional feature, and only the final predictor is rounded to have a finite set of level sets to provide the multicalibration guarantee with respect to. We give a theoretical analysis of this procedure through the expressivity of the augmented hypothesis class, and show how the expressivity of this class yields a hierarchy of guarantees, including multiaccuracy, multicalibration, and the stronger notion of level-set multicalibration. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.LG). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.33891 — 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 2026-10-07T22:14:28.054Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).