ProbXiv
Explorersign in
Problem archiveProblem record

Statement

Let G\mathcal{G} be a library of candidate groups, let x\mathbf{x} be a single observation, let ψ(G,x)\psi(G,\mathbf{x}) be the spectral concentration criterion, and let δ(G,R)\delta(G,\mathbf{R}) be the sample commutativity residual for population covariance R\mathbf{R}. Define the optimal group by

G∗=arg⁡min⁡G∈Gδ(G,R).G^*=\arg\min_{G\in\mathcal{G}}\delta(G,\mathbf{R}).

Blind group selection consistency conjecture. The spectral-concentration selector is consistent at high signal-to-noise ratio:

Pr⁡ ⁣[arg⁡max⁡G∈Gψ(G,x)=G∗]→1as SNR⁡→∞.\Pr\!\left[\arg\max_{G\in\mathcal{G}}\psi(G,\mathbf{x})=G^*\right]\to 1\quad\text{as }\operatorname{SNR}\to\infty.

This would provide a blind, single-snapshot method for selecting the group whose action best matches the covariance, without knowing R\mathbf{R}. The supplied text gives no evidence that the claim has been proved or disproved.

Record

Comments

Nothing has been published against this problem yet, and nothing has been said about it.

Sign in with an institutional address to take part in the discussion. Reading every thread stays open to everyone.

Sign in

Solve with an agent

Open the statement in a chat, with the problem and the ground rules already written into the prompt.

This opens a third-party site. Nothing is posted back to ProbXiv and nothing you write there is recorded here — what a model gives you is an attempt, which a person still has to check.