Id: palomar/PALOMAR-2026-11-09-000006 Url: https://github.com/lopatto/tensor-pca-public/tree/bb8819c75d4c469e58bbe39b30856901751f87cb Json: https://data.palomar-registry.org/entries/PALOMAR-2026-11-09-000006-v1.json Repo: https://github.com/lopatto/tensor-pca-public Tags: math.PR, math.ST, msc:60B20, msc:82B44 Score: 0 Title: Statistical thresholds for Tensor PCA Trust: high Domain: math.PR Source: palomar Authors: Patrick Lopatto Reviews: 0 Subject: math.PR Category: research Comments: 0 Disputed: false Division: math Mentions: 0 Verified: true Discussed: 0 Published: 2026-11-09 Formalized: true Formalizes: title=Statistical thresholds for Tensor PCA — maintainer’s local revision for autoformalization identifier=Statistical thresholds for Tensor PCA, maintainer’s local revision for autoformalization relationship=formalizes Page: https://delphi.science/r/palomar/PALOMAR-2026-11-09-000006 We formalize in Lean the principal statistical and constrained-landscape results of Statistical thresholds for Tensor PCA by Jagannath, Lopatto and Miolane, following a local expanded manuscript prepared solely by Patrick Lopatto, the formalization’s maintainer. This elaborated manuscript was prepared to facilitate autoformalization. At nonnegative signal strength in the ordered Gaussian tensor model, we prove the detection threshold below and above criticality for even orders k ≥ 6. For k ≥ 3, we prove the maximum-likelihood value limit at every nonnegative signal strength and the…