Paper Info
Title
Distributed large-scale tensor decomposition
Abstract
Canonical Polyadic Decomposition (CPD), also known as PARAFAC, is a useful tool for tensor factorization. It has found application in several domains including signal processing and data mining. With the deluge of data faced in our societies, large-scale matrix and tensor factorizations become a crucial issue. Few works have been devoted to large-scale tensor factorizations. In this paper, we introduce a fully distributed method to compute the CPD of a large-scale data tensor across a network of machines with limited computation resources. The proposed approach is based on collaboration between the machines in the network across the three modes of the data tensor. Such a multi-modal collaboration allows an essentially unique reconstruction of the factor matrices in an efficient way. We provide an analysis of the computation and communication cost of the proposed scheme and address the problem of minimizing communication costs while maximizing the use of available computation resources.
Year
DOI
Venue
2014
10.1109/ICASSP.2014.6853551
Acoustics, Speech and Signal Processing
Keywords
Field
DocType
data handling,distributed processing,matrix decomposition,tensors,CPD,PARAFAC,canonical polyadic decomposition,communication cost minimization,computation resources,data mining,distributed computation,distributed large-scale tensor decomposition,factor matrices,large-scale data tensor,large-scale matrix,large-scale tensor factorizations,multimodal collaboration,signal processing,Tensor decompositions,distributed computation,large-scale data
Signal processing,Mathematical optimization,Essentially unique,Tensor,Matrix (mathematics),Computer science,Theoretical computer science,Multilinear subspace learning,Tensor factorization,Computation,Tensor decomposition
Conference
ISSN
Citations 
PageRank 
1520-6149
10
0.53
References 
Authors
19
2
Name
Order
Citations
PageRank
André L. F. de Almeida137148.73
Alain Y. Kibangou29512.01