A Flexible and Efficient Algorithmic Framework for Constrained Matrix and Tensor Factorization
From MaRDI portal
(Redirected from Publication:4620959)
Abstract: We propose a general algorithmic framework for constrained matrix and tensor factorization, which is widely used in signal processing and machine learning. The new framework is a hybrid between alternating optimization (AO) and the alternating direction method of multipliers (ADMM): each matrix factor is updated in turn, using ADMM, hence the name AO-ADMM. This combination can naturally accommodate a great variety of constraints on the factor matrices, and almost all possible loss measures for the fitting. Computation caching and warm start strategies are used to ensure that each update is evaluated efficiently, while the outer AO framework exploits recent developments in block coordinate descent (BCD)-type methods which help ensure that every limit point is a stationary point, as well as faster and more robust convergence in practice. Three special cases are studied in detail: non-negative matrix/tensor factorization, constrained matrix/tensor completion, and dictionary learning. Extensive simulations and experiments with real data are used to showcase the effectiveness and broad applicability of the proposed framework.
Recommendations
- Parallel Algorithms for Constrained Tensor Factorization via Alternating Direction Method of Multipliers
- A constrained matrix factorization problem
- Publication:3468552
- Algorithms for nonnegative matrix and tensor factorizations: a unified view based on block coordinate descent framework
- Tensor decomposition and approximation schemes for constraint satisfaction problems
- Constrained factorization method for nonnegative matrix
- Structured Matrix Approximations via Tensor Decompositions
- Multiplicative algorithms for symmetric nonnegative tensor factorizations and its applications
Cited in
(19)- Hyperspectral image classification based on mathematical morphology and tensor decomposition
- Algorithms for nonnegative matrix factorization with the Kullback-Leibler divergence
- Pairwise constrained concept factorization for data representation
- A survey on deep matrix factorizations
- Exploiting Efficient Representations in Large-Scale Tensor Decompositions
- Hyperspectral Super-resolution Accounting for Spectral Variability: Coupled Tensor LL1-Based Recovery and Blind Unmixing of the Unknown Super-resolution Image
- Best nonnegative rank-one approximations of tensors
- Generalized canonical polyadic tensor decomposition
- Time-homogeneous top-\(K\) ranking using tensor decompositions
- Dynamical variational autoencoders: a comprehensive review
- An AO-ADMM Approach to Constraining PARAFAC2 on All Modes
- Sparsifying the least-squares approach to PCA: comparison of lasso and cardinality constraint
- Tutorial on Amortized Optimization
- Quasi-orthogonalization for alternating non-negative tensor factorization
- Alternating direction method of multipliers for a class of nonconvex bilinear optimization: convergence analysis and applications
- A hybrid and inexact algorithm for nonconvex and nonsmooth optimization
- Efficient Multidimensional Functional Data Analysis Using Marginal Product Basis Systems
- Inertial accelerated stochastic mirror descent for large-scale generalized tensor CP decomposition
- Quantization aware factorization for deep neural network compression
This page was built for publication: A Flexible and Efficient Algorithmic Framework for Constrained Matrix and Tensor Factorization
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4620959)
