2017/06/06 by Takeru Matsuda, Fumiyasu Komaki
Computer Science · Engineering · Mathematics · #Algorithm #Bayes' theorem #Bayesian probability #Blind Source Separation Techniques #Computer science #Estimator #Heuristic #Mathematical optimization #Mathematics #Matrix (chemical analysis) #Sparse and Compressive Sensing Techniques #Statistical Methods and Inference #Statistics #math.ST #stat.ME #stat.ML #stat.TH
paper · pdf · doi:10.1016/j.csda.2019.02.006
published as Computational Statistics & Data Analysis, Vol. 137, pp. 195--210, 2019 · 15 pages
arxiv created 2017/06/06 · openalex publication_date 2019/03/01 · arxiv updated 2019/04/10 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
We develop an empirical Bayes (EB) algorithm for the matrix completion problems. The EB algorithm is motivated from the singular value shrinkage estimator for matrix means by Efron and Morris (1972). Since the EB algorithm is essentially the EM algorithm applied to a simple model, it does not require heuristic parameter tuning other than tolerance. Numerical results demonstrated that the EB algorithm achieves a good trade-off between accuracy and efficiency compared to existing algorithms and that it works particularly well when the difference between the number of rows and columns is large. Application to real data also shows the practical utility of the EB algorithm.