0541614 20250310155920.620210409235959.9 000637533400010 85100948273 10.1109/JSTSP.2021.3059521 10 s. P cav_un_epca*03444611932-4553153 (2021)550-559Institute of Electrical and Electronics Engineers canonical polyadic tensor decomposition parallel factor analysis tensor chain sensitivity cav_un_auth*0101212 Tichavský Petr UTIA-B Stochastická informatika Department of Stochastic Informatics SI SI Department of Stochastic Informatics Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0382249 Phan A. H. RU cav_un_auth*0382250 Cichocki A. RU http://library.utia.cas.cz/separaty/2021/SI/tichavsky-0541614.pdf https://ieeexplore.ieee.org/document/9354901 GA20-17720S GA ČR CZ cav_un_auth*0396617 Structured Tucker tensor decomposition models complete or incomplete multiway data sets (tensors), where the core tensor and the factor matrices can obey different constraints. The model includes block-term decomposition or canonical polyadic decomposition as special cases. We propose a very flexible optimization method for the structured Tucker decomposition problem, based on the second-order Levenberg-Marquardt optimization, using an approximation of the Hessian matrix by the Krylov subspace method. An algorithm with limited sensitivity of the decomposition is included. The proposed algorithm is shown to perform well in comparison to existing tensor decomposition methods. WOS BB 20000 20200 20201 2022 3 2 R hod 4 4rh 4 20250310155350.5 4 20250310155920.6 RVO:67985556 http://hdl.handle.net/11104/0319267 S 2 Article Engineering Electrical Electronic C ENGINEERING.ELECTRICAL&ELECTRONIC 7.992 1.067 5.6 0.01301 2.578 8019 152 3.227 52.34 9.44 3282 Q1 7.606 104 Q1 88.9 Q1* Q1* 2.04 Q1 88.949 2021 Soubory v repozitáři: tichavsky-0541614.pdf 85100948273 SCOPUS PUBMED 000637533400010 WOS cav_un_epca*0344461 IEEE Journal on Selected Topics in Signal Processing 1932-4553 1941-0484 Roč. 15 č. 3 2021 550 559 Institute of Electrical and Electronics Engineers