Matlab implementation of the Scaled Gradient Descent for low-rank tensor estimation under the t-SVD framework (see the JMLR paper link). This code has been tested in the following computational environment.

• PC: Windows 10 with Matlab R2025a

If you use or build upon this work, please cite the following:

@article{Wu.JMLR2026,
    author = {Wu, Tong},
    title = {Guaranteed Nonconvex Low-Rank Tensor Estimation via Scaled Gradient Descent},
	journal = {Journal of Machine Learning Research},
	volume = {27},
	number = {9},
	pages = {1-90},
    year = {2026},
}