🧬Genetic Algorithm

NEW

A population-based metaheuristic that evolves candidate solutions via selection, crossover, and mutation to optimize a fitness function.

DESCRIPTION

Genetic algorithms simulate natural selection on encoded candidate solutions. They iteratively select fitter individuals, recombine genetic material via crossover, and introduce variability through mutation. GAs balance exploration and exploitation and excel at combinatorial and nonconvex optimization. Fitness design, encoding choice, and operator tuning drive performance. They can hybridize with local search to improve convergence. Common applications include scheduling, feature selection, design optimization, and machine learning hyperparameter search.

ALTERNATIVE DEFINITIONS

Wikipedia

A genetic algorithm (GA) is a metaheuristic inspired by the process of natural selection that belongs to the larger class of evolutionary algorithms.

Positions GA within evolutionary algorithms and emphasizes natural selection inspiration.

Source

NIST Dictionary of Algorithms and Data Structures

A genetic algorithm is a search heuristic that mimics the process of natural selection using recombination and mutation of a population of candidate solutions.

Highlights population-based search with recombination and mutation.

Source

MathWorks

The genetic algorithm is a method for solving both constrained and unconstrained optimization problems that is based on natural selection, the process that drives biological evolution.

Emphasizes use for constrained and unconstrained optimization and toolbox context.

Source

MIT OpenCourseWare

Genetic algorithms evolve a population of candidate solutions using selection, crossover, and mutation operators, guided by a fitness measure toward improved performance.

Summarizes the core GA operators and fitness guidance.

Source

RELATED TERMS

Evolutionary Algorithms Filtering Crossover Mutation Fitness Function Population Chromosome Optimization

COMPETING TERMS

Simulated Annealing Particle Swarm Optimization Evolution Strategies

RELATIONS

• Broader: Algorithm

KNOWLEDGE AREAS

AI Mathematics Data Science

TARGET AUDIENCE

Data Scientist AI Engineer Software Developer

EXTERNAL SOURCES

• Wikipedia - Genetic algorithm
• MIT OCW - Lecture on GA fundamentals
• NIST DADS - Concise GA definition
• MathWorks - Toolbox documentation on GA

TOP QUESTIONS

What problems suit a Genetic Algorithm?

Combinatorial optimization, nonlinear continuous optimization, feature selection, and hyperparameter tuning benefit from GA’s exploration and robustness.

Category: Example

How big should the GA population be?

Moderate sizes (50–500) are common; scale based on problem complexity and computational budget.

Category: Requirements

What is a common GA mistake?

Poor encoding and unbalanced mutation/crossover rates can cause premature convergence or slow progress.

Category: Mistakes

Why use crossover in Genetic Algorithms?

Crossover recombines building blocks from fit individuals, accelerating discovery of high-quality solutions.

Category: Example

What risks apply to Genetic Algorithms?

Premature convergence, overfitting to the fitness function, and high computational cost if parameters are poorly tuned.

Category: Risks

How do elitism and selection pressure affect GA?

Elitism preserves best solutions; too much selection pressure reduces diversity and risks premature convergence.

Category: Opinion

Why are Genetic Algorithms still relevant?

They remain strong baselines, integrate well with hybrids, and handle complex landscapes where derivatives are unavailable.

Category: Relevancy

What competes with Genetic Algorithms?

Simulated Annealing, Particle Swarm Optimization, and Evolution Strategies are prominent alternatives.

Category: Alternatives

How to start learning Genetic Algorithms?

Study encoding, fitness design, operator tuning; implement with DEAP or a toolbox; analyze convergence and diversity metrics.

Category: HowTo