Περίληψη :
Modern space- and ground-based surveys are turning astronomy into a data-intensive computational science. Experiments such as DESI, Euclid, and Rubin LSST are mapping the large-scale structure of the Universe at unprecedented scale, producing high-dimensional fields whose scientific value depends on robust inference under complex observational systematics. In this talk, I will describe how modern computational approaches from machine learning and artificial intelligence can enhance cosmological analysis and help answer some of the most important open questions about the nature of our Universe. I will highlight two complementary directions from my work. First, I develop interpretable, multi-scale signal-processing representations that compress nonlinear structure while retaining physical content. Second, I build simulation-based and likelihood-free inference frameworks that go beyond traditional techniques and enable more optimal extraction of cosmological information from these massive datasets. I will conclude with a brief outlook on broader implications for reliable and interpretable AI in astronomy, and on how these ideas motivate open, reusable software that supports interdisciplinary collaboration.
Short Bio: Georgios Valogiannis is a Schmidt AI in Science Fellow in the Department of Astronomy and Astrophysics at the University of Chicago and an affiliate researcher at the NSF–Simons AI Institute for the Sky (SkAI) in the Chicago area. Prior to moving to Chicago, he was a postdoctoral researcher in the Department of Physics at Harvard University, where he worked in Prof. Cora Dvorkin’s group. He received his PhD in Astronomy from Cornell University in 2020. He was born and raised in Greece and earned his BSc in Physics from the Aristotle University of Thessaloniki before moving to the United States in 2014. His research program focuses on developing and applying state-of-the-art artificial intelligence techniques to address some of the most challenging problems in modern astrophysics. In particular, he is interested in using these methods to extract cosmological information from next-generation galaxy surveys, with the goal of shedding light on the nature of dark energy, dark matter, and other fundamental properties of the Universe.
