Climate Resilience and the Role of AI: a Convergence of Complexities

Abstract

Developing climate resilience is a societal grand challenge that intersects the natural, human-engineered, and social dimensions. A broad definition of Artificial Intelligence that includes machine learning or computer vision, graphical or network methods, and agent-based or decision-making models, may contribute to addressing the challenge especially when combined with process models and understanding. However, the nature of the challenge motivates new developments in AI. Machine learning must be augmented with a deeper understanding of nonlinear physics and a better ability to handle extreme values or rare events. Graphical methods must be informed by science and engineering principles including nonlinear relations that may exhibit significant lags in space and time. Agent-based models must be informed by realistic behavior models to encapsulate individual and group dynamics. Among the key considerations are the ability to integrate process-based knowledge with data-driven methods in a principled manner, the need to deal with both Big Data and small data challenges, and a due consideration for the characterization and propagation of uncertainty and variability with a view towards enabling risk-informed decisions.

Biography

Auroop R. Ganguly is a Professor of Civil and Environmental Engineering, a College of Engineering Distinguished Professor, and Co-Director of the Global Resilience Institute, at Northeastern University in Boston, MA, where he has affiliate appointments with the Khoury College of Computer Science and the School of Public Policy and Urban Affairs. His research intersects weather and hydrologic extremes under climate change, lifeline infrastructures resilience under compound extremes, as well as machine learning and nonlinear physics. He has a joint role as a Chief Scientist at the Pacific Northwest National Laboratory. Ganguly co-founded the Boston-based climate analytics startup risQ. Prior to Northeastern, Ganguly worked at the Oak Ridge National Laboratory, a research institute of the US Department of Energy, and at Oracle Corporation, along with a best-of-breed called Demantra Inc. which was subsequently acquired by Oracle. He has been in review panels of the United Nations (UN) Environmental Programme and other US and global agencies, his work has been cited by the UN and US intergovernmental and national reports, and he has delivered invited/keynote talks at workshops organized by the US National Academies and NSF. Ganguly is a Fellow of the American Society of Civil Engineers (ASCE), a Senior Member of the Association for Computing Machinery (ACM) as well as a Senior Member of the Institute of Electrical and Electronics Engineers (IEEE), and obtained a PhD from the Massachusetts Institute of Technology (MIT).

Flip through Auroop’s slides