An optimization technique inspired by natural evolution. It evolves a population of candidate solutions (trading rules, model parameters, etc.) by selection, crossover and mutation to improve performance. In forex robots, GAs can be used to optimize strategy parameters or feature combinations by simulating “survival of the fittest” among strategies. While no single source is cited here, genetic algorithms are common in automated trading for fine-tuning complex systems when gradient-based methods are unsuitable.

Back to Robots Glossary Index