Efficient Nonnegative Tucker Decompositions: Algorithms and Uniqueness

DOI10.1109/TIP.2015.2478396MaRDI QIDQ4613249zbMATH OpenOpenAlexWikidataFDO

Authors Guoxu Zhou, Qibin Zhao, Andrzej Cichocki, Shengli Xie

Publication date 31 January 2019

Published in IEEE Transactions on Image Processing (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1404.4412

Mathematics Subject Classification ID

Image processing (compression, reconstruction, etc.) in information and communication theory (94A08)

Abstract: Nonnegative Tucker decomposition (NTD) is a powerful tool for the extraction of nonnegative parts-based and physically meaningful latent components from high-dimensional tensor data while preserving the natural multilinear structure of data. However, as the data tensor often has multiple modes and is large-scale, existing NTD algorithms suffer from a very high computational complexity in terms of both storage and computation time, which has been one major obstacle for practical applications of NTD. To overcome these disadvantages, we show how low (multilinear) rank approximation (LRA) of tensors is able to significantly simplify the computation of the gradients of the cost function, upon which a family of efficient first-order NTD algorithms are developed. Besides dramatically reducing the storage complexity and running time, the new algorithms are quite flexible and robust to noise because any well-established LRA approaches can be applied. We also show how nonnegativity incorporating sparsity substantially improves the uniqueness property and partially alleviates the curse of dimensionality of the Tucker decompositions. Simulation results on synthetic and real-world data justify the validity and high efficiency of the proposed NTD algorithms.

Cited in

(16)

• A video watermark algorithm based on tensor decomposition
• Multiplicative algorithms for symmetric nonnegative tensor factorizations and its applications
• Randomized algorithms for the low multilinear rank approximations of tensors
• Nonnegative tensor-based linear dynamical systems for recognizing human action from 3D skeletons
• Randomized algorithms for the approximations of Tucker and the tensor train decompositions
• Fast nonnegative tensor factorizations with tensor train model
• Exploiting Efficient Representations in Large-Scale Tensor Decompositions
• The Computation of Low Multilinear Rank Approximations of Tensors via Power Scheme and Random Projection
• Best nonnegative rank-one approximations of tensors
• Sketching for a low-rank nonnegative matrix approximation: numerical study
• Tensor nonconvex unified prior for tensor recovery
• On non-genuinely entangled quantum states by a regularized alternative minimization method
• Tensor train approximation of multivariate Gaussian density by scaling and squaring
• Accurate regularized Tucker decomposition for image restoration
• Graph regularized non-negative bi-level form of triple decomposition for data representation
• Robust low-rank tensor completion based on M-product with weighted correlated total variation and sparse regularization

This page was built for publication: Efficient Nonnegative Tucker Decompositions: Algorithms and Uniqueness

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4613249)