Skip to main content
top

Bibliography

Conference Paper (international conference)

Rank Splitting for CANDECOMP/PARAFAC

Phan A. H., Tichavský Petr, Cichocki A.

: Latent Variable Analysis and Signal Separation, p. 31-40 , Eds: Vincent Emmanuel, Yeredor Arie, Koldovský Zbyněk, Tichavský Petr

: Latent Variable Analysis and Signal Separation 12th International Conference, LVA/ICA 2015, (Liberec, CZ, 25.08.2015-28.08.2015)

: GA14-13713S, GA ČR

: Canonical polyadic decomposition, PARAFAC, deflation

: 10.1007/978-3-319-22482-4_4

: http://library.utia.cas.cz/separaty/2015/SI/tichavsky-0447197.pdf

(eng): CANDECOMP/PARAFAC (CP) approximates multiway data by a sum of rank-1 tensors. Our recent study has presented a method to rank-1 tensor deflation, i.e. sequential extraction of rank-1 tensor components. In this paper, we extend the method to block deflation problem. When at least two factor matrices have full column rank, one can extract two rank-1 tensors simultaneously, and rank of the data tensor is reduced by 2. For decomposition of order-3 tensors of size R×R×R and rank-R, the block deflation has a complexity of O(R^3) per iteration which is lower than the cost O(R^4) of the ALS algorithm for the overall CP decomposition.

: BB