Abstract
In the standard cosmological model, the matter content of the Universe is dominated by dark matter—an invisible component that governs the formation and evolution of cosmic structures. While dark matter cannot be observed directly, its gravitational influence can be detected through the deflection of light from background sources, a phenomenon known as gravitational lensing.
In the first part of this talk, I will introduce the gravitational lensing effect in galaxy clusters and explain how it serves as a powerful probe for detecting dark matter and constraining its physical properties. In the second part, I will present a novel simulation-based inference framework for cluster cosmology. By integrating machine learning with forward modeling of weak lensing observables, this approach bypasses traditional likelihood-based methods and enables robust inference of cosmological parameters.