Institutional Repository [SANDBOX]
Technical University of Crete
EN  |  EL

Search

Browse

My Space

Nonnegative tensor completion: step-sizes for an accelerated variation of the stochastic gradient descent

Liavas Athanasios, Papagiannakos Ioannis-Marios, Kolomvakis Christos

Full record

URI: http://purl.tuc.gr/dl/dias/285CC582-D9A2-4892-A1C7-F42A176CBE38
Year 2022
Type of Item Conference Full Paper
License
Details
Bibliographic Citation A. P. Liavas, I. M. Papagiannakos and C. Kolomvakis, "Nonnegative tensor completion: step-sizes for an accelerated variation of the stochastic gradient descent," in Proceedings of the 30th European Signal Processing Conference (EUSIPCO 2022), Belgrade, Serbia, 2022, pp. 1976-1980, doi: 10.23919/EUSIPCO55093.2022.9909736. https://doi.org/10.23919/EUSIPCO55093.2022.9909736
Appears in Collections
Conference Publications in Community School of Electrical and Computer Engineering
Conference Publications in Community Telecommunications Laboratory

Summary

We consider the problem of nonnegative tensor completion. We adopt the alternating optimization framework and solve each nonnegative matrix least-squares problem via an accelerated variation of the stochastic gradient descent. The step-sizes used by the algorithm determine, to a high extent, its behavior. We propose two new strategies for the computation of step-sizes and we experimentally test their effectiveness using both synthetic and real-world data.

Services

Statistics

 

Copyright © DIAS 2013