Fast Nonnegative Tensor Factorization by Using Accelerated Proximal Gradient.

Nonnegative tensor factorization (NTF) has been widely applied in high-dimensional nonnegative tensor data analysis. However, existing algorithms suffer from slow convergence caused by the non-negativity constraint and hence their practical applications are severely limited. By combining accelerated proximal gradient and low-rank approximation, we propose a new NTF algorithm which is significantly faster than state-of-the-art NTF algorithms.