Fast nonnegative tensor factorization based on accelerated proximal gradient and low-rank approximation

Nonnegative tensor factorization (NTF) has been widely applied in high-dimensional nonnegative tensor data analysis. However, most of the existing algorithms suffer from slow convergence caused by the nonnegativity constraint and hence their practical applications are severely limited. In this study, we propose a new algorithm called FastNTF_APG to speed up NTF by combining accelerated proximal gradient and low-rank approximation. Experimental results demonstrate that FastNTF_APG achieves significantly higher computational efficiency than state-of-the-art NTF algorithms.

[1]  Rasmus Bro,et al.  The N-way Toolbox for MATLAB , 2000 .

[2]  Andrzej Cichocki,et al.  L1-Regularized Multiway Canonical Correlation Analysis for SSVEP-Based BCI , 2013, IEEE Transactions on Neural Systems and Rehabilitation Engineering.

[3]  Bo Du,et al.  Compression of hyperspectral remote sensing images by tensor approach , 2015, Neurocomputing.

[4]  Liqing Zhang,et al.  Multifactor sparse feature extraction using Convolutive Nonnegative Tucker Decomposition , 2014, Neurocomputing.

[5]  Zhigang Luo,et al.  NeNMF: An Optimal Gradient Method for Nonnegative Matrix Factorization , 2012, IEEE Transactions on Signal Processing.

[6]  Tamara G. Kolda,et al.  Tensor Decompositions and Applications , 2009, SIAM Rev..

[7]  Andrzej Cichocki,et al.  Nonnegative Matrix and Tensor Factorization T , 2007 .

[8]  Xingyu Wang,et al.  Frequency Recognition in SSVEP-Based BCI using Multiset Canonical Correlation Analysis , 2013, Int. J. Neural Syst..

[9]  Gordon Wetzstein,et al.  Tensor displays , 2012, SIGGRAPH '12.

[10]  Hai-Long Wu,et al.  A comparison of several trilinear second-order calibration algorithms , 2011 .

[11]  Andrzej Cichocki,et al.  Fast Alternating LS Algorithms for High Order CANDECOMP/PARAFAC Tensor Factorizations , 2013, IEEE Transactions on Signal Processing.

[12]  Xingyu Wang,et al.  SSVEP recognition using common feature analysis in brain–computer interface , 2015, Journal of Neuroscience Methods.

[13]  J. Kruskal.  Three-way arrays: rank and uniqueness of trilinear decompositions, with application to arithmetic complexity and statistics , 1977 .

[14]  Andrzej Cichocki,et al.  Fast Nonnegative Tensor Factorization by Using Accelerated Proximal Gradient , 2014, ISNN.

[15]  Xuelong Li,et al.  General Tensor Discriminant Analysis and Gabor Features for Gait Recognition , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  Liqing Zhang,et al.  Robust Bayesian Tensor Factorization for Incomplete Multiway Data , 2014, ArXiv.

[17]  N. Sidiropoulos,et al.  On the uniqueness of multilinear decomposition of N‐way arrays , 2000 .

[18]  Peng Wu,et al.  Unsupervised Multi-Level Non-Negative Matrix Factorization Model: Binary Data Case , 2012, J. Information Security.

[19]  J. Chang,et al.  Analysis of individual differences in multidimensional scaling via an n-way generalization of “Eckart-Young” decomposition , 1970 .

[20]  Geoffrey E. Hinton,et al.  Visualizing Data using t-SNE , 2008 .

[21]  R. Bro,et al.  Practical aspects of PARAFAC modeling of fluorescence excitation‐emission data , 2003 .

[22]  Gordon Wetzstein,et al.  Tensor displays , 2012, ACM Trans. Graph..

[23]  Hai-Long Wu,et al.  Alternating penalty trilinear decomposition algorithm for second‐order calibration with application to interference‐free analysis of excitation–emission matrix fluorescence data , 2005 .

[24]  Hai-Long Wu,et al.  On the self‐weighted alternating trilinear decomposition algorithm—the property of being insensitive to excess factors used in calculation , 2001 .

[25]  Andrzej Cichocki,et al.  Fast Nonnegative Matrix/Tensor Factorization Based on Low-Rank Approximation , 2012, IEEE Transactions on Signal Processing.

[26]  Pierre Comon,et al.  Fast Decomposition of Large Nonnegative Tensors , 2015, IEEE Signal Processing Letters.

[27]  Tapani Ristaniemi,et al.  Low-rank Approximation Based non-Negative Multi-Way Array Decomposition on Event-Related potentials , 2014, Int. J. Neural Syst..

[28]  Richard A. Harshman,et al.  Foundations of the PARAFAC procedure: Models and conditions for an "explanatory" multi-model factor analysis , 1970 .

[29]  Y. Nesterov.  A method for solving the convex programming problem with convergence rate O(1/k^2) , 1983 .

[30]  Haesun Park,et al.  Fast Nonnegative Matrix Factorization: An Active-Set-Like Method and Comparisons , 2011, SIAM J. Sci. Comput..

[31]  H. Sebastian Seung,et al.  Algorithms for Non-negative Matrix Factorization , 2000, NIPS.