Tuning Free Rank-Sparse Bayesian Matrix and Tensor Completion with Global-Local Priors
arXiv:1905.11496
Abstract
Matrix and tensor completion are frameworks for a wide range of problems, including collaborative filtering, missing data, and image reconstruction. Missing entries are estimated by leveraging an assumption that the matrix or tensor is low-rank. Most existing Bayesian techniques encourage rank-sparsity by modelling factorized matrices and tensors with Normal-Gamma priors. However, the Horseshoe prior and other "global-local" formulations provide tuning-parameter-free solutions which may better achieve simultaneous rank-sparsity and missing-value recovery. We find these global-local priors outperform commonly used alternatives in simulations and in a collaborative filtering task predicting board game ratings.
References in corpus (13)
- A Simpler Approach to Matrix Completion
- Scalable Tensor Factorizations for Incomplete Data
- Bayesian CP Factorization of Incomplete Tensors with Automatic Rank Determination
- A simple sampler for the horseshoe estimator
- Bayesian Robust Tensor Factorization for Incomplete Multiway Data
- Rank regularization and Bayesian inference for tensor completion and extrapolation
- Bayesian methods for low-rank matrix estimation: short survey and theoretical study
- Higher-dimension Tensor Completion via Low-rank Tensor Ring Decomposition
- Bayesian matrix completion: prior specification
- The Inverse Gamma-Gamma Prior for Optimal Posterior Contraction and Multiple Hypothesis Testing
- Matrix factorization for multivariate time series analysis
- Bayesian Matrix Completion via Adaptive Relaxed Spectral Regularization
- Imputation and low-rank estimation with Missing Not At Random data