Constrained CPD of Complex-Valued Multi-Subject fMRI Data via Alternating Rank-R and Rank-1 Least Squares.

"the proposed sar r r 1 ls-po (the code is available in https://github com/lidankuang/cpd_sarrr1ls-po ) can relax cpd model capture large spatiotemporal variabilities and avoid increasing crosstalk between components of noisy complex-valued multi-subject fmri data."

"the proposed sar r r 1 ls-po (the code is available in https://github com/lidankuang/cpd_sarrr1ls-po ) can relax cpd model capture large spatiotemporal variabilities and avoid increasing crosstalk between components of noisy complex-valued multi-subject fmri data."

"This work was supported in part by the National Natural Science Foundation of China under Grant 61901061, Grant 61871067, Grant 61379012, and Grant 61972056; in part by the Natural Science Foundation of Hunan Province under Grant 2020JJ5603; in part by the Research Foundation of Education Bureau of Hunan Province under Grant 19C0031; in part by NSF under Grant 1539067, Grant 0840895, Grant 1539067, and Grant 0715022; in part by NIH under Grant R01MH104680, Grant R01MH107354, Grant R01EB005846, and Grant 5P20GM103472; and in part by the Fundamental Research Funds for the Central Universities, China, under Grant DUT20ZD220."