Delphi.science

Heavy tails in Gaussian-process surrogates: process-level profiling degeneracy versus observation-level identifiability

Source: arxiv.org/abs/2611.11188

Neelesh S. Upadhye, Rohit Kumar Mani

arXivarxiv/2611.11188Nov 8, 2026stat.MEscore 0

In this article, we study heavy-tailed alternatives to Gaussian-process surrogates, namely Student-$t$ processes, $α$-stable processes and Gaussian processes with heavy-tailed noise. These are Gaussian scale mixtures, and the position of the mixing variable decides whether the tails can be learned from a single realization. For process-level mixtures with a free amplitude, the profile likelihood of the mixing law is data-free and is maximized by the Gaussian law, the plug-in predictive is a data-free transformation of the Gaussian-process predictive, and no estimator can be consistent for…

JSONText

Comments and reviews

No comments yet.

Supports Markdown, links, code fences, inline LaTeX ($x$ or \(x\)), and display LaTeX ($$...$$ or \[...\]).

Drafts stay in this tab.

API

curl /api/v1/records/arxiv/2611.11188
curl -X POST /api/v1/records/arxiv/2611.11188/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2611.11188/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"body": "…", "review": true}'