Our Universe is a strange place. All the things we know, from atoms to stars, only constitute about 5% of the Universe’s content. The rest is split among dark matter (~25%) and dark energy (~70%). We believe dark matter should be some kind of elementary particle responsible to shaping up the large-scale structure of the Universe. On the other hand, we know very little about dark energy, besides being responsible of the current accelerated expansion of the Universe.

In the most accepted theoretical model, the Universe started with an explosion called the Big Bang. The fluctuations in the primordial plasma, originating from quantum fluctuations during cosmic inflation, were amplified by gravity; dark matter started clustering in those primordial gravitational potential wells, making them deeper and more massive, which in turn attracted more dark matter and gas to them. The gas fell into these potential wells and became cooler and denser, until stars were formed. This process gave rise to the abundance and spatial distribution of galaxies we observe in the Universe.

This model is able to precisely explain a very large and diverse set of cosmological observations, from the temperature anisotropies in the early Universe to the spatial distribution of galaxies in the present Universe. The model has some free parameters describing fundamental properties of the Universe, such as its age, geometry, composition…etc. Determining the values of these parameters with the highest accuracy is one of the main goals of modern cosmology. Knowing the values of these parameters as precisely as possible will allow us to shed light on some of the most fundamental questions in physics, such as: What is the nature of dark energy? What is the sum of the neutrino masses? How fast is the Universe expanding?

There is a huge amount of information in cosmological observations about the values of the cosmological parameters that is currently not being extracted since the non-linear evolution of cosmic structures gives rise to very complicated, and poorly understood, relationships between observables and these fundamental parameters. Cosmologists usually rely on summary statistics of observed data in order to describe those relationships, but a lot of information is lost in the process.