Constraining Ωm with Phase-Space Information of Galaxies - Natali Soler Matubaro De Santi

Explore the application of astrostatistics and machine learning in constraining the matter density parameter (Ωm) using galaxy phase-space information in this 31-minute conference talk by Natali Soler Matubaro De Santi from Flatiron Institute. Delve into the methodology, including Graph Neural Networks, and examine the data sets used to develop the best model for predictions. Understand why this approach is particularly effective and gain insights into its implications for galaxy formation physics. The presentation covers the introduction, methods, data sets, model performance, results, predictions, and key takeaways, concluding with a Q&A session. This talk is part of a broader conference exploring data-driven tools in galaxy formation and evolution, emphasizing the translation of data-driven results to physical understanding.

Introduction
Method
Graph Neural Networks
Data Sets
Best Model
Best Results
Predictions
Why is it so good
Takeaway
Questions