Algorithmic Foundations for the Diffraction Limit

被引:17
作者
Chen, Sitan [1 ]
Moitra, Ankur [1 ,2 ]
机构
[1] MIT, EECS, Cambridge, MA 02139 USA
[2] MIT, Math, Cambridge, MA 02139 USA
来源
STOC '21: PROCEEDINGS OF THE 53RD ANNUAL ACM SIGACT SYMPOSIUM ON THEORY OF COMPUTING | 2021年
关键词
Mixture models; Fourier optics; matrix pencil method; tensor decomposition; extremal functions; EXTREMAL-FUNCTIONS; SUPERRESOLUTION; MIXTURES;
D O I
10.1145/3406325.3451078
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
For more than a century and a half it has been widely-believed (but was never rigorously shown) that the physics of diffraction imposes certain fundamental limits on the resolution of an optical system. However our understanding of what exactly can and cannot be resolved has never risen above heuristic arguments which, even worse, appear contradictory. In this work we remedy this gap by studying the diffraction limit as a statistical inverse problem and, based on connections to provable algorithms for learning mixture models, we rigorously prove upper and lower bounds on the statistical and algorithmic complexity needed to resolve closely spaced point sources. In particular we show that there is a phase transition where the sample complexity goes from polynomial to exponential. Surprisingly, we show that this does not occur at the Abbe limit, which has long been presumed to be the true diffraction limit.
引用
收藏
页码:490 / 503
页数:14
相关论文
共 48 条
[1]  
Abbe E., 1873, Arch. Mikrosk. Anat., V9, P413, DOI [10.1007/BF02956173, DOI 10.1007/BF02956173]
[2]   On spectral learning of mixtures of distributions [J].
Achlioptas, D ;
McSherry, R .
LEARNING THEORY, PROCEEDINGS, 2005, 3559 :458-469
[3]  
[Anonymous], 1835, Trans. Cambridge Philos. Soc.
[4]  
[Anonymous], 2001, P 33 ANN ACM S THEOR
[5]  
[Anonymous], 1904, An Introduction to the Theory of Optics
[6]   POLYNOMIAL LEARNING OF DISTRIBUTION FAMILIES [J].
Belkin, Mikhail ;
Sinha, Kaushik .
SIAM JOURNAL ON COMPUTING, 2015, 44 (04) :889-911
[7]  
Brubaker SC, 2008, BOLYAI SOC MATH STUD, V19, P241, DOI 10.1007/978-3-540-85221-6_8
[8]  
Buxton A, 1937, PHILOS MAG, V23, P440
[9]   Super-Resolution from Noisy Data [J].
Candes, Emmanuel J. ;
Fernandez-Granda, Carlos .
JOURNAL OF FOURIER ANALYSIS AND APPLICATIONS, 2013, 19 (06) :1229-1254
[10]   Towards a Mathematical Theory of Super- resolution [J].
Candes, Emmanuel J. ;
Fernandez-Granda, Carlos .
COMMUNICATIONS ON PURE AND APPLIED MATHEMATICS, 2014, 67 (06) :906-956