Tensor decompositions for learning latent variable models
From MaRDI portal
Abstract: This work considers a computationally and statistically efficient parameter estimation method for a wide class of latent variable models---including Gaussian mixture models, hidden Markov models, and latent Dirichlet allocation---which exploits a certain tensor structure in their low-order observable moments (typically, of second- and third-order). Specifically, parameter estimation is reduced to the problem of extracting a certain (orthogonal) decomposition of a symmetric tensor derived from the moments; this decomposition can be viewed as a natural generalization of the singular value decomposition for matrices. Although tensor decompositions are generally intractable to compute, the decomposition of these specially structured tensors can be efficiently obtained by a variety of approaches, including power iterations and maximization approaches (similar to the case of matrices). A detailed analysis of a robust tensor power method is provided, establishing an analogue of Wedin's perturbation theorem for the singular vectors of matrices. This implies a robust and computationally tractable estimation approach for several popular latent variable models.
Recommendations
Cited in
(only showing first 100 items - show all)- Complete decomposition of symmetric tensors in linear time and polylogarithmic precision
- On a geometric method for the identifiability of forms
- When random tensors meet random matrices
- Randomized Joint Diagonalization of Symmetric Matrices
- Tensor robust principal component analysis via the tensor nuclear over Frobenius norm
- Asymptotic behavior of large Gaussian correlated Wishart matrices
- Model-Based Tensor Low-Rank Clustering
- Limit behavior in high-dimensional regime for the Wishart tensors in Wiener chaos
- Statistical limits of spiked tensor models
- Low tubal rank tensor recovery using the Bürer-Monteiro factorisation approach. Application to optical coherence tomography
- Eigenconfigurations of tensors
- Tensor train spectral method for learning of hidden Markov models (HMM)
- Three-way clustering of multi-tissue multi-individual gene expression data using semi-nonnegative tensor decomposition
- Tensor decompositions and sparse log-linear models
- Optimal estimation of Gaussian mixtures via denoised method of moments
- The Spacey Random Walk: A Stochastic Process for Higher-Order Data
- Guaranteed Functional Tensor Singular Value Decomposition
- Anomaly detection in scientific data using joint statistical moments
- Learning diagonal Gaussian mixture models and incomplete tensor decompositions
- Large-dimensional independent component analysis: statistical optimality and computational tractability
- Covariate-Assisted Sparse Tensor Completion
- scientific article; zbMATH DE number 5269047 (Why is no real title available?)
- ISLET: fast and optimal low-rank tensor regression via importance sketching
- Convergence rate analysis for the higher order power method in best rank one approximations of tensors
- Clustering a mixture of Gaussians with unknown covariance
- Reducing uncertainty of dynamic heterogeneous information networks: a fusing reconstructing approach
- Lower bounds on the rank and symmetric rank of real tensors
- Smoothed analysis of tensor decompositions
- On the subdifferential of symmetric convex functions of the spectrum for symmetric and orthogonally decomposable tensors
- On polynomial time methods for exact low-rank tensor completion
- A random matrix approach to low-multilinear-rank tensor approximation
- scientific article; zbMATH DE number 7370581 (Why is no real title available?)
- On the number of Waring decompositions for a generic polynomial vector
- The Sup-norm Perturbation of HOSVD and Low Rank Tensor Denoising
- Moment Estimation for Nonparametric Mixture Models through Implicit Tensor Decomposition
- Online tensor methods for learning latent variable models
- Exploiting Efficient Representations in Large-Scale Tensor Decompositions
- Successive Rank-One Approximations for Nearly Orthogonally Decomposable Symmetric Tensors
- Minimality and uniqueness for decompositions of specific ternary forms
- Efficient semiparametric estimation and model selection for multidimensional mixtures
- Orthogonal and unitary tensor decomposition from an algebraic perspective
- The search problem in mixture models
- Robustness and Tractability for Non-convex M-estimators
- Fast moment estimation for generalized latent Dirichlet models
- Learning Markov models via low-rank optimization
- Effective criteria for specific identifiability of tensors and forms
- Statistical Inference, Learning and Models in Big Data
- An operator theoretic approach to nonparametric mixture models
- Learning attribute patterns in high-dimensional structured latent attribute models
- A new method of moments for latent variable models
- Overcomplete order-3 tensor decomposition, blind deconvolution, and Gaussian mixture models
- Provable sparse tensor decomposition
- Spectral methods for learning discrete latent tree models
- An optimal statistical and computational framework for generalized tensor estimation
- Inference for low-rank tensors -- no need to debias
- Iterative hard thresholding for low CP-rank tensor models
- Real eigenstructure of regular simplex tensors
- On the identifiability of ternary forms
- Generalized Low-Rank Plus Sparse Tensor Estimation by Fast Riemannian Optimization
- On uniqueness of power sum decomposition
- Smoothed tensor quantile regression estimation for longitudinal data
- Accelerated distributed expectation-maximization algorithms for the parameter estimation in multivariate Gaussian mixture models
- Successive partial-symmetric rank-one algorithms for almost unitarily decomposable conjugate partial-symmetric tensors
- Tensor decomposition for learning Gaussian mixtures from moments
- Subgroup-effects models for the analysis of personal treatment effects
- Statistically optimal and computationally efficient low rank tensor completion from noisy entries
- Smoothed analysis for tensor methods in unsupervised learning
- The condition number of join decompositions
- The Hitchhiker guide to: secant varieties and tensor decomposition
- Multiresolution low-rank tensor formats
- Tensor-tensor products with invertible linear transforms
- Training Gaussian mixture models at scale via coresets
- High-Dimensional Vector Autoregressive Time Series Modeling via Tensor Decomposition
- Community detection in complex networks: from statistical foundations to data science applications
- Comparison of Accuracy and Scalability of Gauss--Newton and Alternating Least Squares for CANDECOMC/PARAFAC Decomposition
- Non-convex matrix completion and related problems via strong duality
- A moment-matching metric for latent variable generative models
- scientific article; zbMATH DE number 7370576 (Why is no real title available?)
- scientific article; zbMATH DE number 7306859 (Why is no real title available?)
- Leveraging independence in high-dimensional mixed linear regression
- Absolute reconstruction for sums of powers of linear forms: degree 3 and beyond
- A spectral algorithm for latent Dirichlet allocation
- Ergodicity Coefficients for Higher-Order Stochastic Processes
- Learning Topic Models: Identifiability and Finite-Sample Analysis
- Entrywise limit theorems for eigenvectors of signal-plus-noise matrix models with weak signals
- Robust Eigenvectors of Symmetric Tensors
- Polar decomposition-based algorithms on the product of Stiefel manifolds with applications in tensor approximation
- STORE: sparse tensor response regression and neuroimaging analysis
- Tensor-on-tensor regression: Riemannian optimization, over-parameterization, statistical-computational gap and their interplay
- Least squares moment identification of binary regression mixture models
- Robust high-dimensional factor models with applications to statistical machine learning
- Structuring data with block term decomposition: decomposition of joint tensors and variational block term decomposition as a parametrized mixture distribution model
- Lower Memory Oblivious (Tensor) Subspace Embeddings with Fewer Random Bits: Modewise Methods for Least Squares
- Guarantees for existence of a best canonical polyadic approximation of a noisy low-rank tensor
- Estimating Higher-Order Moments Using Symmetric Tensor Decomposition
- Assigning topics to documents by successive projections
- Identifiability for mixtures of centered Gaussians and sums of powers of quadratics
- Newton correction methods for computing real eigenpairs of symmetric tensors
- Method of moments for estimation of noisy curves
- The set of orthogonal tensor trains
This page was built for publication: Tensor decompositions for learning latent variable models
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2934101)
