Efficient Tensor Decomposition with Boolean Factors

Abstract

Tensor decomposition has been extensively used as a tool for exploratory analysis. Motivated by neuroscience applications, we study tensor decomposition with Boolean factors. The resulting optimization problem is challenging due to the non-convex objective and the combinatorial constraints. We propose Binary Matching Pursuit (BMP), a novel generalization of the matching pursuit strategy to decompose the tensor efficiently. BMP iteratively searches for atoms in a greedy fashion. The greedy atom search step is solved efficiently via a MAXCUT-like boolean quadratic program. We prove that BMP is guaranteed to converge sublinearly to the optimal solution and recover the factors under mild identifiability conditions. Experiments demonstrate the superior performance of our method over baselines on synthetic and real datasets. We also showcase the application of BMP in quantifying neural interactions underlying high-resolution spatiotemporal ECoG recordings.

arXiv comment14 pages, 3 figures

References in corpus (9)

• Provable Tensor Factorization with Missing Data
• Convex Tensor Decomposition via Structured Schatten Norm Regularization
• MAD-Bayes: MAP-based Asymptotic Derivations from Bayes
• Gaussian-binary Restricted Boltzmann Machines on Modeling Natural Image Statistics
• The Mixing method: low-rank coordinate descent for semidefinite programming with diagonal constraints
• General Latent Feature Models for Heterogeneous Datasets
• Clustering Boolean Tensors
• Orthogonalized ALS: A Theoretically Principled Tensor Decomposition Algorithm for Practical Use
• Learning Binary Latent Variable Models: A Tensor Eigenpair Approach