As our understanding of the Universe is expanding, the desire to model the physics that govern cosmic evolution is more evident than ever, driving the emergence of cosmological simulations that model the Universe from the beginning of time till present day. In combination with Machine Learning, they allow for an unprecedented capability; one can train AI models on simulations, where the evolution history of galaxies is available, that can in turn be applied on real galaxies. In this work, we propose the use of Python as a ML tool, through the popular library Tensorflow, to quantify the impact of different cosmological models on the derivation of the history of galaxies. Python accompanies us at every step of the way, from creating the datasets and training the probabilistic neural networks to the visualization of the results, as we attempt to shed light on the cosmic past of galaxies, surpassing the unshakeable reality that we can only observe them at a specific moment in time.

This talk is part of our research as a PhD students in the Institute of Astrophysics of the Canary Islands, studying the evolution and formation of galaxies. We would like to present how Python has helped us all the way providing an easy yet powerful ML framework (Tensorflow and tensorflow-probability) and great libraries for scientific processing and visualization (pandas, seaborn etc.).

The key points of the talk will be:

1. Brief introduction to the formation and evolution of galaxies basics and cosmological simulations
2. Using simulation-based inference by training robust probabilistic convolutional neural network models across two cosmological simulations with a domain adaptation scheme
3. Building more complex neural network architectures with Python (e.g. Variational Autoencoders, Normalizing flows)