Centre for Operational Research and Logistics
Events and Seminars
An Efficient Gauss-Newton Algorithm for Symmetric Low-Rank Product Matrix Approximations
Dr Xin Liu will present his recent study of a Gauss-Newton method for computing a symmetric low-rank product XX', where X is an n by n matrix for k<n, that is the closest to a given symmetric matrix in Frobenius norm.
The Gauss-Newton method, which has a particularly simple form, shares the same order of iteration-complexity as a gradient method when k<<n, but can be significantly faster on a wide range of problems. Numerical results show that the proposed algorithm is capable of providing considerable speed advantages over Krylov subspace methods on suitable application problems. Moreover, the algorithm possesses a higher degree of concurrency than Krylov subspace methods, thus offering better scalability on modern multi/many-core computers.
Dr Xin Liu is an Associate Professor at State Key Laboratory of Scientific and Engineering Computing, Institute of Computational Mathematics and Scientific/Engineering Computing in the Academy of Mathematics and Systems Science, Chinese Academy of Sciences, China. He got his BSc on Computational Mathematics from the School of Mathematical Sciences at Peking University in 2004 and PhD on Nonlinear Optimization at the Chinese Academy of Sciences in 2009. He has been a visiting scholar in many places, including Germany, United States, Hong Kong and Singapore. Dr Liu's current research interest is computational methods for optimization, mainly focusing on nonlinear least squares, sparse optimization and matrix decompositions.
Dr Xin Liu, Chinese Academy of Sciences, China
Dennis Sciama Building, DS2.07
Date and Time
Mon, 21 July 2014, 11:00 - 12:00 (BST)
For further information or any enquiries please contact our event co-ordinator, Jana Ries, at email@example.com