PhD student Natasha Randall (https://www.linkedin.com/in/natasha-r-49568a150/) presented her work on generating photorealistic flood predictions with Generative Adversarial Networks (GANs) at the 37th event of the Cologne AI and Machine Learning Meetup (CAIML), which was hosted at TH Köln on July 8, 2025.
In her talk, she explored how GANs can be used to generate flood imagery that is not only realistic, but also provides accurate predictions of the floodwaters resulting from real-world natural disasters, highlighting the benefits of bringing AI into the dynamic fields of hydrology and geosciences.
Natasha does her PhD at TH Köln in cooperation with PK NRW and is working in the Cologne AI and Machine Learning group of Prof. Dr. Heisenberg at TH Köln.
The CAIML event facilitated further networking between the university and machine learning enthusiasts, who attended from a variety of local companies and organizations (https://www.meetup.com/de-DE/cologne-ai-and-machine-learning-meetup).