A partial derandomization of phaselift using spherical designs
From MaRDI portal
Abstract: 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 efficient, 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 we 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. We prove reconstruction guarantees for a number of measurements that depends on the degree t of the design. If the degree is allowed to 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.
Recommendations
Cites work
- Averaging sets: A generalization of mean values and spherical designs
- Chebyshev-type quadrature on multidimensional domains
- Expander graphs and their applications
- Finite Fields and Applications
- Guaranteed minimum-rank solutions of linear matrix equations via nuclear norm minimization
- scientific article; zbMATH DE number 5968745 (Why is no real title available?)
- scientific article; zbMATH DE number 1284419 (Why is no real title available?)
- scientific article; zbMATH DE number 1421280 (Why is no real title available?)
- Immersions and Embeddings of Projective Spaces
- Large deviation bounds for \(k\)-designs
- Low-rank matrix completion using alternating minimization
- On sparse reconstruction from Fourier and Gaussian measurements
- Pairwise Independence and Derandomization
- Quantum invariants of knots and 3-manifolds
- Quantum tomography under prior information
- RIPless compressed sensing from anisotropic measurements
- Sparse signal recovery from quadratic measurements via convex programming
- Spherical 7-designs in \(2^n\)-dimensional Euclidean space
- Structured random measurements in signal processing
- Symmetric informationally complete quantum measurements
- t-designs in projective spaces
- The invariants of the Clifford groups
- Tight informationally complete quantum measurements
- UNITARY OPERATOR BASES
- User-friendly tail bounds for sums of random matrices
Cited in
(39)- Phase retrieval with one or two diffraction patterns by alternating projections with the null initialization
- Infinite-dimensional compressed sensing and function interpolation
- Phase retrieval using random cubatures and fusion frames of positive semidefinite matrices
- Fourier phase retrieval with a single mask by Douglas-Rachford algorithms
- A geometric analysis of phase retrieval
- Phase retrieval from Fourier measurements with masks
- Phase retrieval with PhaseLift algorithm
- Riemannian optimization for phase retrieval from masked Fourier measurements
- Proof methods for robust low-rank matrix recovery
- Phase retrieval using alternating minimization in a batch setting
- Complex phase retrieval from subgaussian measurements
- On the search for tight frames of low coherence
- Algorithms and error bounds for noisy phase retrieval with low-redundancy frames
- Phase retrieval from coded diffraction patterns
- Admissible measurements and robust algorithms for ptychography
- Infinite dimensional compressed sensing from anisotropic measurements and applications to inverse problems in PDE
- Efficient unitary designs with a system-size independent number of non-Clifford gates
- Dynamical quantum tomography
- Breaking the coherence barrier: a new theory for compressed sensing
- Solving Random Quadratic Systems of Equations Is Nearly as Easy as Solving Linear Systems
- Achieving the orthoplex bound and constructing weighted complex projective 2-designs with Singer sets
- Improved recovery guarantees for phase retrieval from coded diffraction patterns
- Low rank matrix recovery from rank one measurements
- Harmonic mean iteratively reweighted least squares for low-rank matrix recovery
- Constrained quantum tomography of semi-algebraic sets with applications to low-rank matrix recovery
- Stable low-rank matrix recovery via null space properties
- Low-Rank Matrix Estimation from Rank-One Projections by Unlifted Convex Optimization
- PhaseMax: stable guarantees from noisy sub-Gaussian measurements
- Well-conditioned ptychograpic imaging via lost subspace completion
- Generalized sampling and infinite-dimensional compressed sensing
- Statistical analysis of compressive low rank tomography with random measurements
- Statistically efficient tomography of low rank states with incomplete measurements
- Fast state tomography with optimal error bounds
- On construction of finite averaging sets for SL(2,C) via its Cartan decomposition
- The numerics of phase retrieval
- Fundamental solutions of the heat equation on unitary groups establish an improved relation between -nets and approximate unitary t-designs
- Robust outlier bound condition to phase retrieval with adversarial sparse outliers
- Explicit frames for deterministic phase retrieval via PhaseLift
- Letter to the editor. Stable low-rank matrix recovery from 3-designs
This page was built for publication: A partial derandomization of phaselift using spherical designs
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2342167)
