Particle Filters for Single-objective Numerical Optimization

This thesis introduces a novel approach combining Particle Filters and the L-BFGS-B optimization method for solving single-objective numerical optimization problems. The proposed method intricately marries the stochastic exploration of Particle Filters with the local optimization prowess of L-BFGS-B to navigate complex landscapes efficiently. Extensive experimentation on benchmark problems validates the approach's effectiveness, convergence speed, accuracy, and robustness. This fusion of methodologies opens new vistas for conquering diverse optimization challenges.