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

2021/03/04 by Ghalamkari, Kazu, Sugiyama, Mahito
#FOS: Computer and information sciences #I.2.6 #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · doi:10.48550/arxiv.2103.02898

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.

Related

• Low Rank Approximation of Tensors via Sparse Optimization
• A literature survey of low-rank tensor approximation techniques
• Practical Sketching Algorithms for Low-Rank Tucker Approximation of Large Tensors
• Nearly Low Rank Tensors and Their Approximations
• Low Rank Tensor Decompositions and Approximations