Genetic Algorithm

A genetic algorithm is an optimization technique inspired by biological evolution that iteratively improves a population of candidate solutions through selection, crossover, and mutation operations. Each candidate solution is encoded as a "chromosome," and a fitness function evaluates how well it solves the target problem. Solutions with higher fitness are more likely to be selected for reproduction, combining their traits through crossover, while random mutations introduce diversity. Genetic algorithms excel at searching large, complex, or non-differentiable solution spaces where gradient descent methods cannot be applied directly. Applications include neural architecture search, hyperparameter tuning, robotics path planning, scheduling optimization, and game strategy evolution. Unlike gradient-based optimization used in deep learning, genetic algorithms do not require the objective function to be differentiable. They belong to the broader family of evolutionary computation methods and are often used alongside machine learning to optimize system design.