top of page

Research

Exploring the nature of Dark Energy with Modified Gravity and Machine Learning

Interests: Cosmology, Dark Energy, Modified Gravity, Machine Learning, Gravitational Waves

Modified Gravity

Current and coming surveys will require sub-percent agreement in theoretical accuracy to test the different cosmological and gravity scenarios, something which can be performed with Boltzmann solvers, i.e. codes that solve the linear evolution of cosmological perturbations. Given the plethora of gravity models, it is crucial to have a standardized unified way to describe all of them and take them into account in a Boltzmann code like CAMB or CLASS. Dark Energy (DE) and Modified Gravity (MG) models, although at a first glance quite dissimilar, are possible to unify them within the same framework. One way to do this is to map the MG models, to linear order, to some DE fluid via the effective fluid approach. I implemented this approach in the widely used cosmic linear anisotropy solving system (CLASS) for f(R) and Horndeski theories obtaining competitive results compared to other MG codes like hi_class. Now I am currently linking this effective fluid approach to MontePython (a Monte Carlo code for cosmological parameter extraction).

Machine Learning

The synergy of theoretical analyses and machine learning (ML) can provide key insights on the nature of gravity and Dark Energy (DE). Even though several DE and MG models have been proposed as candidates for the accelerated expansion, the final picture is not complete. Even more, all recent results in this area suffer from model bias, that is, the results depend strongly on the theory chosen, something that can lead to misleading conclusions regarding the underlying properties of the Universe. For this reason I have followed a more driven data approach using novel ML techniques which are providing innovative approaches to data reconstruction and can capture features of whole classes of theories, without the need to limit the analysis on a specific model. This allows to discriminate among the plethora of MG theories and search for hints of new physics. My main objective is to quantify how the ability of present and future surveys can improve the current cosmological constraints, both at the background and at the perturbation level.

bottom of page