Detecting Model Misspecification in Cosmology with Scale-Dependent Normalizing Flows
This work addresses model validation for cosmology, which is crucial for interpreting data from upcoming surveys, but it appears incremental as it builds on existing methods like normalizing flows and neural networks.
The authors tackled the challenge of validating theoretical models against cosmological data by developing a framework that combines scale-dependent neural summary statistics with normalizing flows to detect model misspecification through Bayesian evidence estimation, demonstrating its application on matter and gas density fields from CAMELS simulations.
Current and upcoming cosmological surveys will produce unprecedented amounts of high-dimensional data, which require complex high-fidelity forward simulations to accurately model both physical processes and systematic effects which describe the data generation process. However, validating whether our theoretical models accurately describe the observed datasets remains a fundamental challenge. An additional complexity to this task comes from choosing appropriate representations of the data which retain all the relevant cosmological information, while reducing the dimensionality of the original dataset. In this work we present a novel framework combining scale-dependent neural summary statistics with normalizing flows to detect model misspecification in cosmological simulations through Bayesian evidence estimation. By conditioning our neural network models for data compression and evidence estimation on the smoothing scale, we systematically identify where theoretical models break down in a data-driven manner. We demonstrate a first application to our approach using matter and gas density fields from three CAMELS simulation suites with different subgrid physics implementations.