Antonios Mamalakis

Antonios Mamalakis is an Assistant Professor of Data Science at the University of Virginia's School of Data Science and also holds a joint appointment as Assistant Professor in the Department of Environmental Sciences. His work bridges environmental science and advanced data science methodologies, focusing on improving climate and hydrological predictions.

Education:

• Ph.D. in Civil and Environmental Engineering, University of California, Irvine
• M.Sc. in Water Resources and Environmental Engineering, University of Patras, Greece
• Diploma in Civil Engineering, University of Patras, Greece

His research focuses on applying machine learning, Bayesian statistics, and explainable AI to environmental challenges such as predicting extreme weather events, understanding climate teleconnections, and advancing causal inference in climate systems. He is particularly interested in enhancing the interpretability and reliability of AI models in geoscientific applications.

The recent publications highlight his expertise in climate dynamics, especially in tropical-extratropical interactions, precipitation predictability, and model evaluation. His work frequently appears in top-tier journals such as Nature Communications, Nature Climate Change, and Geophysical Research Letters, indicating significant contributions to climate and data science.

Scientific Service:

• Associate Editor, Artificial Intelligence for the Earth Systems (American Meteorological Society)

Mamalakis has been recognized for pioneering the use of explainable AI in geosciences during his time as a research scientist at Colorado State University. His research program at UVA continues to develop robust, interpretable data-driven tools for environmental forecasting. Though specific grants and students are not listed, his active lab and publication record suggest ongoing funded research and mentorship.

He leads the Mamalakis Lab, which focuses on developing and applying cutting-edge data science techniques to solve pressing environmental problems, particularly in climate predictability and extreme event modeling.