In this paper we present a novel strategy to solve optimization problems within a hybrid quantum-classical scheme based on quantum annealing, with a particular focus on QUBO problems. The proposed algorithm is based on an iterative structure where the representation of an objective function into the annealer architecture is learned and already visited solutions are penalized by a tabu-inspired search. The result is a heuristic search equipped with a learning mechanism to improve the encoding of the problem into the quantum architecture. We prove the convergence of the algorithm to a global optimum in the case of general QUBO problems. Our technique is an alternative to the direct reduction of a given optimization problem into the sparse annealer graph.