Learning from Binary Multiway Data: Probabilistic Tensor Decomposition and its Statistical Optimality
arXiv:1811.05076
Abstract
We consider the problem of decomposing a higher-order tensor with binary entries. Such data problems arise frequently in applications such as neuroimaging, recommendation system, topic modeling, and sensor network localization. We propose a multilinear Bernoulli model, develop a rank-constrained likelihood-based estimation method, and obtain the theoretical accuracy guarantees. In contrast to continuous-valued problems, the binary tensor problem exhibits an interesting phase transition phenomenon according to the signal-to-noise ratio. The error bound for the parameter tensor estimation is established, and we show that the obtained rate is minimax optimal under the considered model. Furthermore, we develop an alternating optimization algorithm with convergence guarantees. The efficacy of our approach is demonstrated through both simulations and analyses of multiple data sets on the tasks of tensor completion and clustering.
arXiv comment35 pages, 7 figures, 4 tables
References in corpus (12)
- Tensor Regression with Applications in Neuroimaging Data Analysis
- A modern maximum-likelihood theory for high-dimensional logistic regression
- Square Deal: Lower Bounds and Improved Relaxations for Tensor Recovery
- Matrix Completion has No Spurious Local Minimum
- Generalized Canonical Polyadic Tensor Decomposition
- A Max-Norm Constrained Minimization Approach to 1-Bit Matrix Completion
- Sparse logistic principal components analysis for binary data
- A statistical model for tensor PCA
- Learning tensors from partial binary measurements
- Operator Norm Inequalities between Tensor Unfoldings on the Partition Lattice
- Non-Convex Projected Gradient Descent for Generalized Low-Rank Tensor Regression
- Tensor Decompositions via Two-Mode Higher-Order SVD (HOSVD)