Generative AI Modelling for Extreme Events Extreme events—such as market crashes, wildfires, unprecedented flooding, intense hurricanes, and extreme heatwaves—cause major disruptions across societal and ecological systems.This workshop aims to stimulate discussion on generative AI models for extreme events and to promote the exploration of novel directions in this emerging area.Extreme Value Theory (EVT) provides a robust mathematical framework, drawing on Karamata’s theory of regular variation and asymptotic principles, to estimate the risks of such events by extrapolating into the tails of a distribution—beyond the limits of available data. While particular emphasis on this workshop will be placed on EVT-inspired approaches, a multidisciplinary outlook will be strongly encouraged.The origins of AI can be traced back to the 1920s and 1930s, with foundational contributions from Gödel and Turing in logic and theoretical computer science; interestingly, EVT emerged in a similar period, shaped by pioneering work from Fisher and Tippett. Despite these parallel beginnings, meaningful intersections between the two fields have only recently begun to emerge. This workshop seeks to contribute to this growing convergence, with a particular emphasis on aspects related with generative AI for extremes.Chair: Miguel de Carvalho; Chair of Statistical Data Science and GAIL Fellow Partners This event is funded and organized under the auspices of the Generative AI Laboratory. We gratefully acknowledge the partnership with the Royal Society of Edinburgh, the Edinburgh Centre for Financial Innovations, and the Glasgow–Edinburgh Extremes Network. Jun 13 2025 - Jun 14 2025 Generative AI Modelling for Extreme Events This workshop aims to stimulate discussion on generative AI models for extreme events and to promote the exploration of novel directions in this emerging area. Yew Lecture Theatre, Nucleus Building, King's Buildings, University of Edinburgh Yew Lecture Theatre Register
Generative AI Modelling for Extreme Events Extreme events—such as market crashes, wildfires, unprecedented flooding, intense hurricanes, and extreme heatwaves—cause major disruptions across societal and ecological systems.This workshop aims to stimulate discussion on generative AI models for extreme events and to promote the exploration of novel directions in this emerging area.Extreme Value Theory (EVT) provides a robust mathematical framework, drawing on Karamata’s theory of regular variation and asymptotic principles, to estimate the risks of such events by extrapolating into the tails of a distribution—beyond the limits of available data. While particular emphasis on this workshop will be placed on EVT-inspired approaches, a multidisciplinary outlook will be strongly encouraged.The origins of AI can be traced back to the 1920s and 1930s, with foundational contributions from Gödel and Turing in logic and theoretical computer science; interestingly, EVT emerged in a similar period, shaped by pioneering work from Fisher and Tippett. Despite these parallel beginnings, meaningful intersections between the two fields have only recently begun to emerge. This workshop seeks to contribute to this growing convergence, with a particular emphasis on aspects related with generative AI for extremes.Chair: Miguel de Carvalho; Chair of Statistical Data Science and GAIL Fellow Partners This event is funded and organized under the auspices of the Generative AI Laboratory. We gratefully acknowledge the partnership with the Royal Society of Edinburgh, the Edinburgh Centre for Financial Innovations, and the Glasgow–Edinburgh Extremes Network. Jun 13 2025 - Jun 14 2025 Generative AI Modelling for Extreme Events This workshop aims to stimulate discussion on generative AI models for extreme events and to promote the exploration of novel directions in this emerging area. Yew Lecture Theatre, Nucleus Building, King's Buildings, University of Edinburgh Yew Lecture Theatre Register
Jun 13 2025 - Jun 14 2025 Generative AI Modelling for Extreme Events This workshop aims to stimulate discussion on generative AI models for extreme events and to promote the exploration of novel directions in this emerging area.
