Metraipolis › archive › dc_101ae22e0953
When Does Dynamic Ensembling Pay Off? Diagnosing Regionwise Gains in Regression under Distribution Shift — arXiv:2608.18330
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.18330
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T13:53:13.639Z
- SHA-256 of the archived text
a2f68192d9e34eafdd4f8cd1c242f5e9f1b5c167d07a2fa1ccd7baa3b3c1c1ec- Stable URL of this page
https://metraipolis.com/documento/dc_101ae22e0953
How to cite
When Does Dynamic Ensembling Pay Off? Diagnosing Regionwise Gains in Regression under Distribution Shift — arXiv:2608.18330 — arXiv (en), https://arxiv.org/abs/2608.18330 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_101ae22e0953 (Metraipolis archive dc_101ae22e0953, sha256 a2f68192d9e34eafdd4f8cd1c242f5e9f1b5c167d07a2fa1ccd7baa3b3c1c1ec)
BibTeX
@misc{dc101ae22e0953,
title = {When Does Dynamic Ensembling Pay Off? Diagnosing Regionwise Gains in Regression under Distribution Shift — arXiv:2608.18330},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.18330},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_101ae22e0953; sha256 a2f68192d9e34eafdd4f8cd1c242f5e9f1b5c167d07a2fa1ccd7baa3b3c1c1ec},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.18330. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Whether input-dependent ("dynamic") combination of a regression model pool beats the best static blend depends on the shift and is rarely known before deployment. Can a small labeled target-domain probe tell us when reallocating trust across regions of the input space will pay off? We answer this with $\widehat{D}_{\mathrm{CF5}}$, which estimates from the probe the cross-fitted gain of the regionwise convex combination over the best static convex blend: the realizable value of deciding, region by region, whom to trust. Across a frozen suite of 12 dataset-shift pairs (spatial, temporal, domain, feature-cluster), $\widehat{D}_{\mathrm{CF5}}$ predicts realized regionwise test gains with dataset-level Spearman $+0.98$ (95% CI $[+0.83, +1.00]$; $p=5\times10^{-5}$), including two cases overturning preregistered expectations. The relationship holds in a 16-pair sensitivity analysis (Spearman $+0.83$), whereas alternative probe diagnostics reach at most $+0.66$. This contrast isolates regional trust reallocation: correlation is $+0.98$ for regionwise-convex gain, but $+0.01$ for smooth covariate-dependent stacking after affine correction. A controlled generator shows dynamic gains arise from the interaction of shift heterogeneity and local competence, increase with shift severity, and become realizable between 128 and 256 probe labels in the tested grid. The Probe-Validated Ensemble Selector chooses among a static affine stacker and dynamic realizers, deploying a candidate only when a held-out lower confidence bound clears the static-convex floor. In a preregistered prospective batch, it matched or improved the floor in all 12 runs; two deployments reduced test risk by 11% and 16%, while the gate rejected a candidate whose un-gated deployment incurred $>30\times$ the static loss. We release OpenRegShift, a reproducible evaluation harness for regression ensembles under distribution shift.
Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-20 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.18330 — 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 2026-10-18T13:50:58.901Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-20?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).