 |
 |
* Title: Sensitivity analysis, Uncertainty quantification and Parameter Estimation of Complex Biological and Environmental Systems * Date: 04/29/2016 * Time: 4:10 PM - 5:00 PM * Place: C304 WH * Speaker: Guang Lin, Purdue University There are many uncertainties in modeling of complex biological and environmental systems. Experience suggests that uncertainties often play an important role in quantifying the performance of complex systems. Therefore, uncertainty needs to be treated as a core element in modeling, simulation and optimization of complex systems. The field of uncertainty quantification (UQ) has received an increasing amount of attention recently. Extensive research efforts have been devoted to it and many novel numerical techniques have been developed. These techniques aim to conduct stochastic simulations for data-driven large-scale complex systems. In this talk, we will present some effective new ways of dealing with the “curse of dimensionality” and “parameter estimation” challenges. Particularly, inverse regression-based uncertainty quantification algorithm and compressive sensing techniques will be discussed in some detail. First, the inverse regression algorithm will be introduced to identify the intrinsically low dimensional subspace, known as the sufficient dimension reduction subspace in the high-dimensional UQ problem. We will illustrate the main idea of our developed dimension reduction UQ algorithm using a groundwater flow in an aquifer problem. Second, we studied the sensitivity analysis and parameter estimation of an Ebola disease epidemic model in Sierra Leone to improve the model predictivity of the Ebola disease spread and outbreaks. In addition, we demonstrate how to use the calibrated Ebola disease epidemic model to investigate various forecasting and control policy assessment for policy making to effectively control and prevent the outbreaks of Ebola disease.