Research Colloquium — Wednesday, October 5, 2005

"Efficient Markov chain Monte Carlo methods for dynamic data integration"

Prof. Yalchin Efendiev
Department of Mathematics, Texas A&M University


Approaches employed in inverse problems related to subsurface characterization often have stochastic nature due to intrinsic uncertainties. In this talk, I will describe the use of Markov chain Monte Carlo (MCMC) methods for solving these inverse problems. I will give a brief overview of MCMC and discuss the advantages and disadvantages of MCMC methods. Then I will talk about our recent work which is designed to improve the standard MCMC methods.



Room: 126 Clements Hall
Coffee: 3:15 pm – 3:30 pm
Colloquium: 3:30 pm – 4:30 pm