Randomized online CP decomposition
Publication Type
Conference Proceeding Article
Publication Date
3-2018
Abstract
CANDECOMP/PARAFAC (CP) decomposition has been widely used to deal with multi-way data. For real-time or large-scale tensors, based on the ideas of randomized-sampling CP decomposition algorithm and online CP decomposition algorithm, a novel CP decomposition algorithm called randomized online CP decomposition (ROCP) is proposed in this paper. The proposed algorithm can avoid forming full Khatri-Rao product, which leads to boost the speed largely and reduce memory usage. The experimental results on synthetic data and real-world data show the ROCP algorithm is able to cope with CP decomposition for large-scale tensors with arbitrary number of dimensions. In addition, ROCP can reduce the computing time and memory usage dramatically, especially for large-scale tensors.
Keywords
CP decomposition, Online learning, Randomized-sampling, Tensor decomposition
Discipline
Databases and Information Systems | Programming Languages and Compilers
Publication
Proceedings of 10th International Conference on Advanced Computational Intelligence, Fujian, China, 2018 March 29-31
First Page
414
Last Page
419
ISBN
9781538643624
Identifier
10.1109/ICACI.2018.8377495
Publisher
Institute of Electrical and Electronics Engineers Inc.
City or Country
City Hotel XiamenXiamen, Fujian, China
Citation
MA, Congbo; YANG, Xiaowei; and WANG, Hu.
Randomized online CP decomposition. (2018). Proceedings of 10th International Conference on Advanced Computational Intelligence, Fujian, China, 2018 March 29-31. 414-419.
Available at: https://ink.library.smu.edu.sg/sis_research/4112
Additional URL
https://doi.org/10.1109/ICACI.2018.8377495