Metraipolis › archive › dc_d0fa4ff7568c
Information-Geometric Blind Spots of Conformal Test Martingales — arXiv:2610.23131
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.23131
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T09:21:22.866Z
- SHA-256 of the archived text
03ad05887fed7290a5225c79dd1b44e11aec2dc8bee5be6c1115b89da015f0e1- Stable URL of this page
https://metraipolis.com/documento/dc_d0fa4ff7568c
How to cite
Information-Geometric Blind Spots of Conformal Test Martingales — arXiv:2610.23131 — arXiv (en), https://arxiv.org/abs/2610.23131 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d0fa4ff7568c (Metraipolis archive dc_d0fa4ff7568c, sha256 03ad05887fed7290a5225c79dd1b44e11aec2dc8bee5be6c1115b89da015f0e1)
BibTeX
@misc{dcd0fa4ff7568c,
title = {Information-Geometric Blind Spots of Conformal Test Martingales — arXiv:2610.23131},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.23131},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d0fa4ff7568c; sha256 03ad05887fed7290a5225c79dd1b44e11aec2dc8bee5be6c1115b89da015f0e1},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.23131. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Conformal test martingales (CTMs) are e-processes for sequentially testing exchangeability via p-values produced by a conformity function $A$. However, any distribution shift that preserves the push-forward $P\circ A^{-1}$ leaves the p-value distribution invariant, rendering it invisible to every CTM based on $A$. We formalise this blind spot through the \emph{$A$-equivalence class} $[P]_A$ and show that, for any target distribution $T$, there is a unique closest element $Q^\ast\in[P]_A$ in KL divergence---the \emph{$A$-cryptic I-projection}---with explicit density $q^\ast(z) = t(z)\,p_S(A(z))/t_S(A(z))$ and a Pythagorean decomposition into detectable and cryptic components. When $A$ is the Neyman-Pearson likelihood ratio, the cryptic component vanishes, connecting our result to optimal e-value design. We demonstrate the theory on high-dimensional image data by empirically constructing this I-projection. By conditionally resampling EMNIST letters to identically match the conformity scores of MNIST digits, we demonstrate that a distribution can be shifted to a disjoint support while leaving the test martingale mathematically blinded.
Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (stat.ML) ; Machine Learning (cs.LG); Statistics Theory (math.ST).
Pagina del lavoro: https://arxiv.org/abs/2610.23131 — 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 2026-10-19T09:17:41.708Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).