Title: Integrating multiple random sketches of singular value decomposition
Authors: Su-Yun Huang - Academia Sinica (Taiwan) [presenting]
Abstract: Low-rank singular value decomposition (SVD) of large-scale matrices is a key tool in modern data analysis and scientific computing. Rapid growing in matrix size further increases the needs and poses the challenges for developing efficient large-scale SVD algorithms. Random sketching is a promising method to reduce the problem size for computing an approximate SVD. We generalize the one-time sketching to multiple random sketches and develop algorithms to integrate these random sketches containing various subspace information in different randomizations. Such integration procedure can lead to SVD with higher accuracy and the multiple randomizations can be conducted on parallel computers simultaneously. We also reveal the insights and analyze the performance of the proposed algorithms from statistical and geometric viewpoints. Numerical results are presented and discussed to demonstrate the efficiency of the proposed algorithms.