Journal Club presentation on "Interpreting Cosmological Information from Neural Networks in the Hydrodynamic Universe". This talk details how neural networks extract cosmological information from hydrodynamic density fields, exploring how CNNs marginalize over complex baryonic effects to constrain cosmological parameters.

Contributed talk at the DAOISM workshop on ISM studies with data science and PPV spectral cubes, held at the Institut d'Astrophysique de Paris (IAP), France. Presenting high-z interferometric denoising benchmarks and physical property inference using deep learning and sparsity.