A partial derandomization of phaselift using spherical designs (Q2342167)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | A partial derandomization of phaselift using spherical designs |
scientific article |
Statements
A partial derandomization of phaselift using spherical designs (English)
0 references
11 May 2015
0 references
The problem of retrieving phase information from amplitude measurements alone has appeared in many scientific disciplines over the last century. PhaseLift is a recently introduced algorithm for phase recovery that is computationally tractable, numerically stable, and comes with rigorous performance guarantees. PhaseLift is optimal in the sense that the number of amplitude measurements required for phase reconstruction scales linearly with the dimension of the signal. However, it specifically demands Gaussian random measurement vectors -- a limitation that restricts practical utility and obscures the specific properties of measurement ensembles that enable phase retrieval. Here the authors present a partial derandomization of PhaseLift that only requires sampling from certain polynomial size vector configurations, called t-designs. Such configurations have been studied in algebraic combinatorics, coding theory, and quantum information. The authors prove reconstruction guarantees for a number of measurements that depends on the degree t of the design. If the degree is allowed to grow logarithmically with the dimension, the bounds become tight up to polylog-factors. Beyond the specific case of PhaseLift, this work highlights the utility of spherical designs for the derandomization of data recovery schemes.
0 references
phase retrieval
0 references
phaselift
0 references
semidefinite relaxations of nonconvex quadratic programs
0 references
non-commutative large deviation estimates
0 references
spherical designs
0 references
convex relaxation
0 references
quantum information
0 references