Speaker
Description
Cold dark matter, the standard cosmological model, faces several challenges on small scales that self-interacting dark matter (SIDM) may help resolve. Traditional methods to constrain the SIDM cross-section often rely on summary statistics, which discard much of the available information, or require complex and computationally expensive lensing models. Machine learning (ML) has gained traction in astronomy for its ability to extract features from high-dimensional data, but its black box nature raises concerns for scientific inference.
We present an interpretable ML algorithm for constraining the SIDM cross-section from cosmological simulations of galaxy clusters. Our algorithm embeds weak lensing maps into a low-dimensional feature space based on their similarity, allowing simulations to cluster based on their physical differences. This feature space provides a way to assess whether a test dataset, such as observations, lies within the training domain, indicating whether any SIDM constraint is reliable or requires extrapolation.
Our ML algorithm provides accurate parameter recovery alongside a measure of prediction confidence for improved ML trustworthiness.