The iterated local search (ILS) is exceptionally successful in combinatorial solution spaces. However, few research works have reported on the application of ILS in continuous problems. In this article, a new hybrid population-based iterated local search (HILS) algorithm is proposed for solving numerical optimization problems. The proposed hybrid method introduces both success-history based parameter adaptation for differential evolution (SHADE) and limited-memory Broyden–Fletcher–Goldfarb–Shanno (LBFGS) as the perturbation and local search strategy, respectively, and integrates the benefits of exploration capability of SHADE and local search performance of LBFGS. The simulated annealing type of acceptance criterion is adopted to balance the exploration and exploitation of ILS. The proposed HILS is tested against the CEC2017 benchmark functions, which were used to evaluate the performance of the proposed algorithm in solving numerical optimization problems. The experimental results show the effectiveness and efficiency of the HILS.