Скачать или смотреть Christopher Wikle. Hybrid Statistical/AI Models for Spatio-Temporal Data: Wildland Fire Applications

Christopher K. Wikle, Curators’ Distinguished Professor of Statistics at U. Missouri (MU)

Hybrid Statistical/AI Models for Spatio-Temporal Data: Wildland Fire Applications

Abstract:

Millions of acres of land are destroyed by wildfires every year, and these fires pose a significant threat to humans both in terms of property damage and loss of life, as well as significantly impacting the ecosystem. It is important to develop trustworthy models that can be used to manage and mitigate large wildfires. Spatio-temporal data such as those associated with wildfires are ubiquitous in the sciences, medicine, and engineering, and their study is important for understanding and predicting a wide variety of processes. One of the difficulties with statistical modeling of spatial processes that change in time is the complexity of the dependence structures that must describe how such a process varies, and the presence of high-dimensional complex datasets and large prediction domains. A perfect example is modeling the evolution of a wildfire, which has many factors contributing to its growth and spread. It has long been the case that deep (hierarchical) statistical models have proven helpful for such data, yet these models can be difficult to implement for various reasons. Incorporating mechanistic processes within the hierarchical modeling framework has proven helpful. Increasingly, black-box neural (“AI”) models are being used for spatio-temporal data as well, capitalizing the strength of those models to learn complex dependence structures. The downside of such models is the requirement for large amounts of training data, interpretability, and uncertainty quantification. It is natural to consider hybrid models that address some of these issues. Here, after a brief background on spatio-temporal modeling in statistics, I present several brief examples of hybrid statistical/AI models for predicting the growth and spread of wildfire.

More about Christopher K. Wikle

Christopher K. Wikle is Curators’ Distinguished Professor of Statistics at U. Missouri (MU), with additional appointments in Soil, Environmental and Atmospheric Sciences and the Truman School of Public Affairs. He obtained his Ph.D. from Iowa State University in 1996 and has been on the faculty at MU for 27 years. His research specialty is spatio-temporal statistics, with applications to geophysical processes, complex biological processes, and the environment. He focuses on developing computationally efficient deep hierarchical Bayesian dynamic spatio-temporal models motivated by scientific principles, with more recent work at the interface of deep neural modeling and statistics. He is Fellow of the ASA, IMS, ISI, and AAAS and has published 2 award winning books in spatio-temporal statistics. Dr. Wikle is Associate Editor for several journals and is one of six inaugural members of the Statistics Board of Reviewing Editors for the AAAS flagship journal, Science.

Webpage: https://wiklec.mufaculty.umsystem.edu/