Speaker
Dr
Konstantinos Dialektopoulos
(University of Malta)
Description
The standard cosmological model is under increasing observational pressure. Persistent tensions in the Hubble constant and the growth of structure suggest either unaccounted systematics or genuine new physics beyond ΛCDM — and distinguishing between these possibilities demands data analysis methods that do not smuggle in theoretical priors through the back door. In this talk, I present a programme of model-independent reconstruction of late-time cosmology using machine learning to probe both the expansion history of the Universe and the underlying theory of gravity.
