☆ 4.5 Article

Practical alternating least squares for tensor ring decomposition

NUMERICAL LINEAR ALGEBRA WITH APPLICATIONS (2023)

Related references

Note: Only part of the references are listed.

Article Mathematics, Applied

CP decomposition for tensors via alternating least squares with QR decomposition

Rachel Minster et al.

Summary: The CP tensor decomposition is commonly used in machine learning and signal processing to discover latent low-rank structure in multidimensional data. This paper proposes new versions of the CP-ALS algorithm that utilize the QR decomposition and the singular value decomposition to solve the linear least squares problems, which are more numerically stable than the traditional normal equations. The algorithms show improved stability and lower approximation error in ill-conditioned cases.

NUMERICAL LINEAR ALGEBRA WITH APPLICATIONS (2023)

Add to Collection

Review Computer Science, Artificial Intelligence

Randomized algorithms for fast computation of low rank tensor ring model

Salman Ahmadi-Asl et al.

Summary: Randomized algorithms are efficient techniques for analyzing big data tensors, particularly in Tensor Ring format. The focus is on using random projection technique for decomposing large-scale data tensors in TR format. Simulations and performance comparisons are provided to support the presentation and efficiency of the algorithms.

MACHINE LEARNING-SCIENCE AND TECHNOLOGY (2021)

Add to Collection

Article Mathematics, Applied