This talk presents a particle-based optimization method designed for addressing global optimization problems, particularly in cases where the loss function exhibits non-differentiability or non-convexity. Numerically, we show that it outperforms gradient-based method in finding global optimizer. Theoretically, A rigorous mean-field limit of the particle system is derived, and the convergence of the mean-field limit to the global minimizer is established. In addition, we will talk about its application to the constrained optimization problems and federated learning.