The noncommutative Gaussian norm conjecture

Let X=k=1sgkAkX=\sum_{k=1}^s g_kA_k be the Gaussian matrix model from the noncommutative Khintchine theorem, where gkg_k are independent standard Gaussian variables and AkA_k are the associated matrices. Let BB be the unit ball used in that theorem, and write X\|X\| for the relevant operator norm. Noncommutative Gaussian norm conjecture.

EXEX21/2+supvBE[v,Xv2]1/2logn.\mathbf{E}\|X\|\lesssim \|\mathbf{E}X^2\|^{1/2}+\sup_{v\in B}\mathbf{E}[\langle v,Xv\rangle^2]^{1/2}\sqrt{\log n}.

This is explicitly described as speculative and would extend the nearly optimal independent-entry bounds to the dependent setting of noncommutative Khintchine inequalities. The source emphasizes that independence is crucial in the existing proofs and gives no evidence of a resolution.

Sources & referencesView supporting material

Primary source

Ramon van Handel, “Structured Random Matrices”, arXiv:1610.05200 (2016).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.