An environmental data scientist, Antonios Mamalakis is interested in using data science tools like statistical and Bayesian analysis, machine/deep learning, and explainable AI to solve challenges in environmental applications. Among others, these challenges include improving predictive skill of hydroclimate and extreme events, understanding climate teleconnections and predictability, advancing climate attribution and causal discovery.

Before joining UVA’s faculty, he worked as a research scientist at Colorado State University, where Mamalakis pioneered the investigation of the fidelity of explainable AI tools for applications in the geosciences. Examples of his papers that garnered international attention and have been highlighted by publishers include "A new interhemispheric teleconnection increases predictability of winter precipitation in southwestern US", published in Nature Communications; "Zonally contrasting shifts of the tropical rain belt in response to climate change", published in Nature Climate Change; "Underestimated MJO variability in CMIP6 models", published in Geophysical Research Letters; and "Climate-driven changes in the predictability of seasonal precipitation", published in Nature Communications. Mamalakis also serves as an associate editor for the AMS journal Artificial Intelligence for the Earth Systems. 

Antonios holds a Ph.D. in civil and environmental engineering from the University of California, Irvine and a master’s in the same discipline from Greece’s University of Patras.