Deep-learning accelerated optimization solvers with convergence guarantees - 42Papers