In this seminar, I will present the main research topics that my supervisors and I have been working on during my Ph.D., and which lie at the intersection of Machine Learning (ML) and Mathematical Programming (MP). The main contributions tackle the Algorithm Configuration Problem (ACP) and the Distance Geometry Problem (DGP). The ACP is the problem of configuring a target algorithm with the parameter setting providing optimal algorithmic performance, for solving a given input. We propose two novel approaches to the ACP, both based on a two-fold process combining ML and MP techniques, and employ them to configure the IBM ILOG CPLEX optimization solver. The DGP consists in finding a realization of a simple, undirected, weighted graph, in a Euclidean space of given dimension, where the edges are realized as straight segments of length equal to the edge weights. A customary approach to the DGP is to solve a MP to determine the position of the vertices in the given Euclidean space. We propose a new MP formulation where, instead, we consider the cycles of the graph, and we decide the length of the segments modelling the edges of each cycle.