A Doubly-Enhanced EM Algorithm for Model-Based Tensor Clustering

Abstract

Modern scientific studies often collect data sets in the forms of tensors, which call for innovative statistical analysis methods. In particular, there is a pressing need for tensor clustering methods to understand the heterogeneity in the data. We propose a tensor normal mixture model (TNMM) approach to enable probabilistic interpretation and computational tractability. Our statistical model leverages the tensor covariance structure to reduce the number of parameters for parsimonious modeling, and at the same time explicitly exploits the correlations for better variable selection and clustering. We propose a doubly-enhanced expectation-maximization (DEEM) algorithm to perform clustering under this model. Both the E-step and the M-step are carefully tailored for tensor data in order to account for statistical accuracy and computational cost in high dimensions. Theoretical studies confirm that DEEM achieves consistent clustering even when the dimension of each mode of the tensors grows at an exponential rate of the sample size. Numerical studies demonstrate favorable performance of DEEM in comparison to existing methods.

References in corpus (12)

• Tensor Regression with Applications in Neuroimaging Data Analysis
• High-dimensional classification using features annealed independence rules
• On Tensors, Sparsity, and Nonnegative Factorizations
• Tensor-on-tensor regression
• Covariance pattern mixture models for the analysis of multivariate heterogeneous longitudinal data
• Separable factor analysis with applications to mortality data
• Provable Convex Co-clustering of Tensors
• Characterizing Spatiotemporal Transcriptome of Human Brain via Low Rank Tensor Decomposition
• Discriminating sample groups with multi-way data
• Singularity, Misspecification, and the Convergence Rate of EM
• Regularized matrix data clustering and its application to image analysis
• Clustering Higher Order Data: An Application to Pediatric Multi-variable Longitudinal Data