Randomized algorithms for the computation of multilinear rank-(μ1,μ2,μ3) approximations

被引：0

作者：

Che, Maolin ^{[1
]}

Wei, Yimin ^{[2
,3
]}

Xu, Yanwei ^{[4
]}

机构：

[1] Southwestern Univ Finance & Econ, Sch Math, Chengdu 611130, Peoples R China

[2] Fudan Univ, Sch Math Sci, Shanghai 200433, Peoples R China

[3] Fudan Univ, Key Lab Math Nonlinear Sci, Shanghai 200433, Peoples R China

[4] Huawei Technol Co Ltd, Theory Lab, Cent Res Inst, 2012 Labs, Hong Kong, Peoples R China

来源：

JOURNAL OF GLOBAL OPTIMIZATION | 2023年 / 87卷 / 2-4期

关键词：

Randomized algorithms; Random projection; Multilinear rank-(mu(1); mu(2); mu(3)); approximation; Sparse subspace embedding; Singular value decomposition; Singular values; The power scheme; TENSOR; TUCKER; MATRIX; DECOMPOSITION;

D O I：

10.1007/s10898-022-01182-8

中图分类号：

C93 [管理学]; O22 [运筹学];

学科分类号：

070105 ; 12 ; 1201 ; 1202 ; 120202 ;

摘要：

We present some randomized algorithms for computing multilinear rank-(mu(1),mu(2),mu(3)) approximations of tensors by combining the sparse subspace embedding and the singular value decomposition. The error bound for this algorithm with the high probability is obtained by the properties of sparse subspace embedding. Furthermore, combining the power scheme and the proposed randomized algorithm, we derive a three-stage randomized algorithm and make a probabilistic analysis for its error bound. The efficiency of the proposed algorithms is illustrated via numerical examples.

引用

页码：373 / 403

页数：31

共 50 条

[21] The Best Rank- (R1, R2, R3) Approximation of Tensors by Means of a Geometric Newton Method
Ishteva, Mariya
De Lathauwer, Lieven
Absil, P. -A.
Van Huffel, Sabine
NUMERICAL ANALYSIS AND APPLIED MATHEMATICS, 2008, 1048 : 274 - +
[22] OPTIMIZATION-BASED ALGORITHMS FOR TENSOR DECOMPOSITIONS: CANONICAL POLYADIC DECOMPOSITION, DECOMPOSITION IN RANK-(Lr, Lr, 1) TERMS, AND A NEW GENERALIZATION
Sorber, Laurent
Van Barel, Marc
De Lathauwer, Lieven
SIAM JOURNAL ON OPTIMIZATION, 2013, 23 (02) : 695 - 720
[23] Differential-geometric Newton method for the best rank-(R1, R2, R3) approximation of tensors
Mariya Ishteva
Lieven De Lathauwer
P.-A. Absil
Sabine Van Huffel
Numerical Algorithms, 2009, 51 : 179 - 194
[24] Efficient Computation of Robust Weighted Low-Rank Matrix Approximations Using the L1 Norm
Eriksson, Anders
van den Hengel, Anton
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2012, 34 (09) : 1681 - 1690
[25] Pattern recognition framework based on the best rank-(R1, R2, ... , RK) tensor approximation
Cyganek, B.
COMPUTATIONAL VISION AND MEDICAL IMAGE PROCESSING IV, 2014, : 301 - 306
[26] Low tensor train and low multilinear rank approximations of 3D tensors for compression and de-speckling of optical coherence tomography images
Kopriva, Ivica
Shi, Fei
Lai, Mingyig
Stanfel, Marija
Chen, Haoyu
Chen, Xinijan
PHYSICS IN MEDICINE AND BIOLOGY, 2023, 68 (12):
[27] A Krylov-Schur-like method for computing the best rank-(r1,r2,r3) approximation of large and sparse tensors
Elden, Lars
Dehghan, Maryam
NUMERICAL ALGORITHMS, 2022, 91 (03) : 1315 - 1347
[28] THE COMPUTATION OF F(3)2(1)
WIMP, J
INTERNATIONAL JOURNAL OF COMPUTER MATHEMATICS, 1981, 10 (01) : 55 - 62
[29] Efficient Computation of Robust Low-Rank Matrix Approximations in the Presence of Missing Data using the L1 Norm
Eriksson, Anders
van den Hengel, Anton
2010 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2010, : 771 - 778
[30] A Krylov-Schur-like method for computing the best rank-(r1,r2,r3) approximation of large and sparse tensors
Lars Eldén
Maryam Dehghan
Numerical Algorithms, 2022, 91 : 1315 - 1347

← 1 2 3 4 5 →