題目：Particle Filter and Its Application to Dynamic Data Driven Simulation
報告人：Dr. Xiaolin Hu, Department of Computer Science, Georgia State University
Assimilating real time sensor data into a running simulation model is an essential component of dynamic data driven simulation, where a simulation system is continually influenced by real time data for better analysis and prediction of a system under study. The complexity of large-scale simulation models asks for sophisticated inference techniques to estimate the distribution or evolution of system states. In this talk, I will present Sequential Monte Carlo (SMC) methods, also called particle filters, for dynamic data driven simulation. SMC methods are a set of sample-based methods that use Bayesian inference and stochastic sampling techniques to recursively estimate the states of dynamic systems from some given observations. They approximate the sequence of probability distributions of interest using a large set of random samples, named particles, and continuously adjust the weights of these particles based on the observation data feedback. A key advantage of SMC methods is their ability to represent arbitrary probability densities, and thus can be applied to systems with complex nonlinear, non-Gussian, unsteady behaviors. I will discuss the foundations of and give a tutorial on SMC methods. Applications of SMC methods to dynamic data driven simulation will also be discussed and demonstrated.
Dr. Xiaolin Hu is an Associate Professor in the Department of Computer Science at Georgia State University, Atlanta, Georgia. He received his Ph.D. degree from the University of Arizona in 2004. His research interests include modeling and simulation theory and application, agent and multi-agent systems, and complex systems science. He has served as program chairs for several international conferences/symposiums in the field of modeling and simulation, and is associate editors for Simulation: Transaction of The Society for Modeling and Simulation International, International Journal of Modeling, Simulation, and Scientific Computing (IJMSSC), and International Journal of Agent Technologies and Systems (IJATS). Dr. Hu is a National Science Foundation (NSF) CAREER Award recipient.