Stochastic Search Methods for Global Optimization

Abstract Stochastic search methods, also known as random search algorithms, are popular for ill‐structured global optimization problems because they are straightforward to implement and usually find a relatively good solution quickly. These algorithms have been inspired by physics, such as simulated annealing and interacting particle algorithms, as well as by biology, including genetic algorithms, evolutionary programming, particle swarm, and ant colony optimization. This article highlights the use of a Markov chain Monte Carlo sampling method called Hit‐and‐Run in random search algorithms for global optimization.