Low-rank nonnegative tensor approximation via alternating projections and sketching

Abstract

We show how to construct nonnegative low-rank approximations of nonnegative tensors in Tucker and tensor train formats. We use alternating projections between the nonnegative orthant and the set of low-rank tensors, using STHOSVD and TTSVD algorithms, respectively, and further accelerate the alternating projections using randomized sketching. The numerical experiments on both synthetic data and hyperspectral images show the decay of the negative elements and that the error of the resulting approximation is close to the initial error obtained with STHOSVD and TTSVD. The proposed method for the Tucker case is superior to the previous ones in terms of computational complexity and decay of negative elements. The tensor train case, to the best of our knowledge, has not been studied before.

Accepted version

References in corpus (1)

• A Parallel Low-Rank Solver for the Six-Dimensional Vlasov-Maxwell Equations

Cited by in corpus (6)

• Sketching for low-rank nonnegative matrix approximation: Numerical study
• Nonnegative tensor train for the multicomponent Smoluchowski equation
• Entrywise tensor-train approximation of large tensors via random embeddings
• Quasioptimal alternating projections and their use in low-rank approximation of matrices and tensors
• When big data actually are low-rank, or entrywise approximation of certain function-generated matrices
• Estimates for the quantized tensor train ranks for the power functions