Almost-everywhere uniqueness conjecture for complex phase retrieval least squares

Let ACm×dA\in \mathbb{C}^{m\times d} have the phase retrieval property in Cd\mathbb{C}^d, meaning that AA determines every vector in Cd\mathbb{C}^d up to the relevant phase ambiguity from its measurement magnitudes. For bRm{\bm b}\in\mathbb{R}^m, consider the nonlinear least-squares program

minxCdAxb22.\min_{{\bm x}\in\mathbb{C}^d}\|\lvert A{\bm x}\rvert-{\bm b}\|_2^2.

Almost-everywhere uniqueness conjecture. For all vectors bRm{\bm b}\in\mathbb{R}^m except for a measure zero set, this program has a unique solution. The real analogue is established by the preceding theorem; the conjectured complex analogue concerns uniqueness for almost every real data vector.

Sources & referencesView supporting material

Primary source

Meng Huang and Zhiqiang Xu, “Uniqueness and stability for the solution of a nonlinear least squares problem”, arXiv:2104.10841 (2021).

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.