Invited Session Thu.3.H 0107

Thursday, 15:15 - 16:45 h, Room: H 0107

Cluster 16: Nonlinear programming [...]

Algorithms and applications II

 

Chair: Ya-xiang Yuan

 

 

Thursday, 15:15 - 15:40 h, Room: H 0107, Talk 1

Jinyan Fan
Accelerating the modified Levenberg-Marquardt method

 

Abstract:
In this talk, we will present an accelerated version of the modified LM method for nonlinear equations, which not only computes a LM step but also an approximate LM step with line search at every iteration. Under the local error bound condition which is weaker than nonsingularity,
we show the convergence order of the accelerated modified LM method is a continuous function with respect to the LM parameter. We compare the new method with both the LM method and the modified LM method and observe significantly competitive performance.

 

 

Thursday, 15:45 - 16:10 h, Room: H 0107, Talk 2

Yanfei Wang
Optimizing inversion methods for seismic imaging

 

Abstract:
In this talk we address several migration and optimizing inversion methods in seismic imaging. In particular, regularizing least squares migration and inversion imaging techniques are discussed. Preconditioning technique is also introduced. Numerical tests are made to show the performance of the methods. Since the interferometric migration and the least squares migration both aim to improve the resolution of seismic imaging, a numerical experiment is also made to discuss their ability in improving imaging resolution.

 

 

Thursday, 16:15 - 16:40 h, Room: H 0107, Talk 3

Torsten Falko Bosse
Limited memory updating and quadratic overestimation for NLOP

Coauthors: Levis Eneya, Andreas Griewank

 

Abstract:
We pursue the approach of solving non-linearly constraint problems using an active set strategy without the complete evaluation of Jacobians or Hessians. The linearized KKT systems are solved approximately on the basis of limited derivative information and compact storage schemes.
The resulting step corresponds to a projected Hessian who's positive definiteness is ensured with the help of quadratic overestimation.
We will present theoretical arguments and numerical experiments.

 

  There are three major facts that should be watched out for in all payday loans in the United States. Of course, the choice is not that easy, as there exist great number of different preparations. Notwithstanding, Cialis is the one that definitely differs from all other products.