Generalized canonical polyadic tensor decomposition
From MaRDI portal
Abstract: Tensor decomposition is a fundamental unsupervised machine learning method in data science, with applications including network analysis and sensor data processing. This work develops a generalized canonical polyadic (GCP) low-rank tensor decomposition that allows other loss functions besides squared error. For instance, we can use logistic loss or Kullback-Leibler divergence, enabling tensor decomposition for binary or count data. We present a variety statistically-motivated loss functions for various scenarios. We provide a generalized framework for computing gradients and handling missing data that enables the use of standard optimization methods for fitting the model. We demonstrate the flexibility of GCP on several real-world examples including interactions in a social network, neural activity in a mouse, and monthly rainfall measurements in India.
Recommendations
- Stochastic gradients for large-scale tensor decomposition
- Tensor Decompositions and Applications
- Spectral learning on matrices and tensors
- Optimization-based algorithms for tensor decompositions: canonical polyadic decomposition, decomposition in rank-(L_r,L_r,1) terms, and a new generalization
- Tensor decomposition and its applications in statistics
Cites work
- A Flexible and Efficient Algorithmic Framework for Constrained Matrix and Tensor Factorization
- A Limited Memory Algorithm for Bound Constrained Optimization
- Algorithm 862
- Algorithms for nonnegative matrix factorization with the -divergence
- Algorithms for Numerical Analysis in High Dimensions
- Analysis of individual differences in multidimensional scaling via an \(n\)-way generalization of ``Eckart-Young decomposition
- Efficient MATLAB Computations with Sparse and Factored Tensors
- Families of alpha-, beta- and gamma-divergences: flexible and robust measures of similarities
- Generalized low rank models
- Learning the parts of objects by non-negative matrix factorization
- Low complexity damped Gauss-Newton algorithms for CANDECOMP/PARAFAC
- Multivariate regression and machine learning with sums of separable functions
- Newton-based optimization for Kullback-Leibler nonnegative tensor factorizations
- Numerical operator calculus in higher dimensions
- On Tensors, Sparsity, and Nonnegative Factorizations
- Positive tensor factorization
- Randomized alternating least squares for canonical tensor decompositions: application to a PDE with random data
- Robust Estimation of a Location Parameter
- Tensor Decompositions and Applications
- Tensor-product approximation to operators and functions in high dimensions
Cited in
(40)- A proximal point like method for solving tensor least-squares problems
- On large-scale dynamic topic modeling with nonnegative CP tensor decomposition
- Equivalent polyadic decompositions of matrix multiplication tensors
- Alternate algorithms to most referenced techniques of numerical optimization to solve the symmetric rank-\(R\) approximation problem of symmetric tensors
- Inertial accelerated SGD algorithms for solving large-scale lower-rank tensor CP decomposition problems
- An optimal statistical and computational framework for generalized tensor estimation
- Tensor factorization recommender systems with dependency
- MuLOT: multi-level optimization of the canonical polyadic tensor decomposition at large-scale
- Generalized co-clustering analysis via regularized alternating least squares
- A nonlinear GMRES optimization algorithm for canonical tensor decomposition
- New Uniqueness Conditions for the Canonical Polyadic Decomposition of Third-Order Tensors
- Coupled Canonical Polyadic Decompositions and Multiple Shift Invariance in Array Processing
- On Koopman mode decomposition and tensor component analysis
- Multiplicative Updates for NMF with $\beta$-Divergences under Disjoint Equality Constraints
- Stochastic gradients for large-scale tensor decomposition
- Generalized Tensor Decomposition With Features on Multiple Modes
- Randomized Projection for Rank-Revealing Matrix Factorizations and Low-Rank Approximations
- Learning from binary multiway data: probabilistic tensor decomposition and its statistical optimality
- Software for Sparse Tensor Decomposition on Emerging Computing Architectures
- Half-quadratic alternating direction method of multipliers for robust orthogonal tensor approximation
- Accelerated doubly stochastic gradient descent for tensor CP decomposition
- Exponential family tensor completion with auxiliary information
- Guaranteed Functional Tensor Singular Value Decomposition
- CP decomposition and low-rank approximation of antisymmetric tensors
- Coseparable nonnegative tensor factorization with t-CUR decomposition
- Robust Low-Rank Tensor Decomposition with the L 2 Criterion
- Joint Tensor Modeling of Single Cell 3D Genome and Epigenetic Data with Muscle
- An alternating shifted higher order power method based algorithm for rank-R Hermitian approximation and solving Hermitian CP-decomposition problems
- Tensor decomposition with unaligned observations
- On non-genuinely entangled quantum states by a regularized alternative minimization method
- A composite optimization algorithm for Poisson tensor completions without nonnegative constraints based on the generalized Gauss-Newton method
- Tensor decompositions for count data that leverage stochastic and deterministic optimization
- Inertial accelerated stochastic mirror descent for large-scale generalized tensor CP decomposition
- Tensor neural network and its numerical integration
- Poisson tensor completion via nonconvex regularization and nonlocal self-similarity for multi-dimensional image recovery
- On the maximal ranks of some third-order quaternion tensors
- A quasi-subspace iteration method for canonical polyadic decomposition to third order tensors
- Multi-step inertial accelerated doubly stochastic gradient methods for block term tensor decomposition
- An improved inertial stochastic proximal alternating linearized minimization for nonconvex optimization in machine learning
- Dimensionality reduction and topographic mapping of binary tensors
This page was built for publication: Generalized canonical polyadic tensor decomposition
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5216247)
