Fast Tucker Rank Reduction for Non-Negative Tensors Using Mean-Field Approximation

Kazu Ghalamkari, Mahito Sugiyama

Keywords: Low-rank approximation, Tucker rank, tensor, mean-field theory, Information geometry

Abstract: We present an efficient low-rank approximation algorithm for non-negative tensors. The algorithm is derived from our two findings: First, we show that rank-1 approximation for tensors can be viewed as a mean-field approximation by treating each tensor as a probability distribution. Second, we theoretically provide a sufficient condition for distribution parameters to reduce Tucker ranks of tensors; interestingly, this sufficient condition can be achieved by iterative application of the mean-field approximation. Since the mean-field approximation is always given as a closed formula, our findings lead to a fast low-rank approximation algorithm without using a gradient method. We empirically demonstrate that our algorithm is faster than the existing non-negative Tucker rank reduction methods and achieves competitive or better approximation of given tensors.

Code Of Conduct: I certify that all co-authors of this work have read and commit to adhering to the NeurIPS Statement on Ethics, Fairness, Inclusivity, and Code of Conduct.

TL;DR: A novel fast Tucker rank reduction method based on mean-field approximation for non-negative tensors

Code: https://github.com/gkazunii/Legendre-tucker-rank-reduction