A new numerical model of the lattice Boltzmann method utilizing least-squares finite element in space and Crank-Nicolson method in time is presented. The new method is able to solve problem domains that contain complex or irregular geometric boundaries by using finite-element method's geometric flexibility and numerical stability, while employing efficient and accurate least-squares optimizatio...

In the machine learning field, feature selection is used to discard the redundant information and improve the learning accuracy. In this paper, the redundant information is reused in the learning of partial least squares method within the frame of multitask learning. This newly proposed method is used to solve the multivariate calibration problem, a classic problem in the analytical chemistry f...

To my beloved wife, Yuijn iii ACKNOWLEDGEMENTS First, I want to thank my wife Yujin for her enduring patience and loving support during my study. She has been always there for me from the very beginning to this very end. She has been at the core of my motivation even when I was doubting if could finish this program. I also have to mention my beloved children, Lynn and Olivia, who have been an e...

in this paper, we propose an algorithm for numerical solving an inverse non-linear diusion problem. in additional, the least-squares method is adopted to nd the solution. to regularize the resultant ill-conditioned linear system ofequations, we apply the tikhonov regularization method to obtain the stablenumerical approximation to the solution. some numerical experiments con- rm the utility of...

In this paper, we first get further consideration of the order perturbation with normwise condition number MTLS problem. For easy estimation, show a lower bound for which is proved to be optimal. overcome problems encountered in calculating number, give an upper computing more effectively and nonstandard unusual bounds Both two types can enjoy storage computational advantages. getting insight i...