矩阵填充算法的比较研究

Matrix Completion concerns how to recover a matrix from a sampling of its entries.There are many application,such as recommender systems, estimation of the heart rate,predicting gene-disease and so on.The recommender systems competition of Netflix Prize make the matrix completion algorithm be a hot topic. Candes and Rechat put forward theconvex optimization theory of matrix completion, provide theoretical framework for currently popular convex optimization to solve matrix completion . Based on the theoretical analysis , performance and pratical application in this paper,analysis the advances of matrix completion and existing problem, and points out the direction of further research.