Novel algorithms for strongly correlated quantum systems in and out of equilibrium

What do stars in a galaxy, drops in a river, and electrons in a superconducting cuprate levitating above a magnet all have in common? All of these systems cannot be described by the isolated motion of one of their parts. These singular properties emerge from particles and their interactions as a whole: we talk about the emph{many-body problem}.In this Thesis, we focus on properties of strongly-correlated systems, that obey quantum mechanics. Analytical methods being rapidly limited in their understanding of these materials, we develop novel numerical techniques to precisely quantify their properties when interactions between particles become strong.First, we focus on the equilibrium properties of the layered perovskite Sr$_2$IrO$_4$, a compound isostructural to the superconducting cuprate La$_2$CuO$_4$,where we prove the existence of a pseudogap and describe the electronic structure of this material upon doping.Then, in order to address the thermodynamic limit of lattice problems, we develop extensions of determinant Monte Carlo algorithms to compute dynamical quantities such as the self-energy. We show how a factorial number of diagrams can be regrouped in a sum of determinants, hence drastically reducing the fermionic sign problem.In the second part, we turn to the description of nonequilibrium phenomena in correlated systems.We start by revisiting the real-time diagrammatic Monte Carlo recent advances in a new basis where all vacuum diagrams directly vanish.In an importance sampling procedure,such an algorithm can directly addressthe long-time limit needed in the study of steady states in out-of-equilibrium systems.Finally, we study the insulator-to-metal transition induced by an electric field in Ca$_2$RuO$_4$, which coexists with a structural transition.An algorithm based on the non-crossing approximation allows us to compute the current as a function of crystal-field splitting in this material and to compare our results to experimental data.