Towards Sample-Optimal Methods for Solving Random Quadratic Equations with Structure

We consider the problem of estimating a structured high-dimensional parameter vector using random Gaussian quadratic samples. This problem is a generalization of the classical problem of phase retrieval and impacts numerous problems in computational imaging. We provide a generic algorithm based on alternating minimization that, if properly initialized, achieves information-theoretically optimal sample complexity. In essence, we show that solving a system of random quadratic equations with structural constraints is (nearly) as easy as solving the corresponding linear system with the same constraints, if a proper initial guess of the solution is available. As an immediate consequence, our approach improves upon the best known existing sample complexity results for phase retrieval (structured or otherwise). We support our theory via several numerical experiments. A full version of this paper is accessible at: https://gaurijagatap.github.io/assets/ISIT18.pdf

[1]  Volkan Cevher,et al.  Model-Based Compressive Sensing , 2008, IEEE Transactions on Information Theory.

[2]  Sjoerd Dirksen,et al.  Tail bounds via generic chaining , 2013, ArXiv.

[3]  Emmanuel J. Candès,et al.  Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information , 2004, IEEE Transactions on Information Theory.

[4]  J R Fienup,et al.  Phase retrieval algorithms: a comparison. , 1982, Applied optics.

[5]  Gang Wang,et al.  Solving Random Systems of Quadratic Equations via Truncated Generalized Gradient Flow , 2016, NIPS.

[6]  Roman Vershynin,et al.  Introduction to the non-asymptotic analysis of random matrices , 2010, Compressed Sensing.

[7]  Mahdi Soltanolkotabi,et al.  Structured Signal Recovery From Quadratic Measurements: Breaking Sample Complexity Barriers via Nonconvex Optimization , 2017, IEEE Transactions on Information Theory.

[8]  Allen Y. Yang,et al.  CPRL -- An Extension of Compressive Sensing to the Phase Retrieval Problem , 2012, NIPS.

[9]  Xiaodong Li,et al.  Phase Retrieval via Wirtinger Flow: Theory and Algorithms , 2014, IEEE Transactions on Information Theory.

[10]  M. Talagrand.  The Generic chaining : upper and lower bounds of stochastic processes , 2005 .

[11]  M. Yuan,et al.  Model selection and estimation in regression with grouped variables , 2006 .

[12]  Yang Wang,et al.  Robust sparse phase retrieval made easy , 2014, 1410.5295.

[13]  Piotr Indyk,et al.  A fast approximation algorithm for tree-sparse recovery , 2014, 2014 IEEE International Symposium on Information Theory.

[14]  Xiaodong Li,et al.  Optimal Rates of Convergence for Noisy Sparse Phase Retrieval via Thresholded Wirtinger Flow , 2015, ArXiv.

[15]  Prateek Jain,et al.  Phase Retrieval Using Alternating Minimization , 2013, IEEE Transactions on Signal Processing.

[16]  Deanna Needell,et al.  Greedy signal recovery review , 2008, 2008 42nd Asilomar Conference on Signals, Systems and Computers.

[17]  R. Gerchberg.  A practical algorithm for the determination of phase from image and diffraction plane pictures , 1972 .

[18]  Gang Wang,et al.  Sparse Phase Retrieval via Truncated Amplitude Flow , 2016, IEEE Transactions on Signal Processing.

[19]  Yuxin Chen,et al.  Solving Random Quadratic Systems of Equations Is Nearly as Easy as Solving Linear Systems , 2015, NIPS.

[20]  Piotr Indyk,et al.  Fast Algorithms for Structured Sparsity , 2015, Bull. EATCS.

[21]  Deanna Needell,et al.  CoSaMP: Iterative signal recovery from incomplete and inaccurate samples , 2008, ArXiv.

[22]  Xiaodong Li,et al.  Phase Retrieval from Coded Diffraction Patterns , 2013, 1310.3240.

[23]  Coralia Cartis,et al.  An Exact Tree Projection Algorithm for Wavelets , 2013, IEEE Signal Processing Letters.

[24]  Gang Wang,et al.  Solving Most Systems of Random Quadratic Equations , 2017, NIPS.

[25]  Richard G. Baraniuk,et al.  Compressive phase retrieval , 2007, SPIE Optical Engineering + Applications.

[26]  Piotr Indyk,et al.  Nearly Linear-Time Model-Based Compressive Sensing , 2014, ICALP.

[27]  Roi Livni,et al.  On the Computational Efficiency of Training Neural Networks , 2014, NIPS.

[28]  Chinmay Hegde,et al.  Sample-Efficient Algorithms for Recovering Structured Signals From Magnitude-Only Measurements , 2017, IEEE Transactions on Information Theory.

[29]  David P. Woodruff,et al.  Lower bounds for sparse recovery , 2010, SODA '10.

[30]  Yingbin Liang,et al.  Reshaped Wirtinger Flow for Solving Quadratic System of Equations , 2016, NIPS.

[31]  Chinmay Hegde,et al.  Fast, Sample-Efficient Algorithms for Structured Phase Retrieval , 2017, NIPS.

[32]  Emmanuel J. Candès,et al.  PhaseLift: Exact and Stable Signal Recovery from Magnitude Measurements via Convex Programming , 2011, ArXiv.