Nonnegative Tucker decomposition with alpha-divergence

Nonnegative Tucker decomposition (NTD) is a recent multiway extension of nonnegative matrix factorization (NMF), where nonnegativity constraints are incorporated into Tucker model. In this paper we consider α-divergence as a discrepancy measure and derive multiplicative updating algorithms for NTD. The proposed multiplicative algorithm includes some existing NMF and NTD algorithms as its special cases, since α-divergence is a one-parameter family of divergences which accommodates KL-divergence, Hellinger divergence, χ2 divergence, and so on. Numerical experiments on face images show how different values of α affect the factorization results under different types of noise.

• Type: Conference
• Year: 2008
• Where: ICASSP
• Authors: Yong-Deok Kim, Andrzej Cichocki, Seungjin Choi
• Keywords: ICASSP 2008, Multiplicative Updating Algorithms, Nonnegative Matrix Factorization, Nonnegative Tucker Decomposition, Signal Processing