In this article, a hybrid inversion algorithm based on an innovative stochastic algorithm, namely, the bat algorithm (BA) is proposed. Electromagnetic inverse scattering problems are ill-posed and are often transformed into optimization problems by defining a suitable cost function. As typical methods to solve optimization problems, stochastic optimization algorithms are more flexible and have better global searching ability than deterministic algorithms. However, they share a common disadvantage: heavy computing load. This directly restricts the application of the algorithms in high-dimensional problems and real-time imaging environments. To solve this issue, diffraction tomography (DT) is introduced to provide a reference for the initialization of the BA. Furthermore, the hybrid method makes full use of the complementary advantages of linear reconstruction algorithms and stochastic optimization algorithms to improve accuracy and efficiency at the same time. Moreover, in order to avoid the algorithm falling into local extrema, a linear attenuation strategy of the pulse emission rate is proposed to enable more bats to perform global search in the early stage of the algorithm. In the numerical experiments for different types of dielectric objects, the reconstruction results of this hybrid BA-based algorithm are compared with those of the DT and the particle swarm optimization (PSO).