TY  - INPR
ID  - SisLab4003
UR  - https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4003/
A1  - Le Trung, Thanh
A1  - Karim, Abed-Meraim
A1  - Nguyen Linh, Trung
A1  - Remy, Boyer
Y1  - 2021/01//
N2  - Tensor-train (TT) decomposition has been an ef?cient tool to ?nd low order approximation of large-scale, high-order tensors. Existing TT decomposition algorithms are either of high computational complexity or operating in batch-mode, hence quite inef?cient for (near) real-time processing. In this paper, we propose a novel adaptive algorithm for TT decomposition of streaming tensors whose slices are serially acquired over time. By leveraging the alternating minimization framework, our estimator minimizes an exponentially weighted least-squares cost function in an ef?cient way. The proposed method can yield an estimation accuracy very close to the error bound. Numerical experiments show that the proposed algorithm is capable of adaptive TT decomposition with a competitive performance evaluation on both synthetic and real data.
TI  - Adaptive Algorithms for Tracking Tensor-Train Decomposition of Streaming Tensors
M2  - Amsterdam, The Netherlands
AV  - none
T2  - 28th European Signal Processing Conference (EUSIPCO)
ER  -