Scribd
Upload
24 views2 pages

Genetic Algorithms - qn-2c

Genetic Algorithms (GA) are optimization tools inspired by natural evolution, utilizing mechanisms like population, chromosomes, genes, fitness functions, selection, crossover, and mutation to efficiently explore complex solution spaces. They enhance local search methods by simultaneously evaluating multiple solutions, maintaining diversity, and iterating through generations to improve overall fitness. GAs offer advantages over traditional local search by increasing parallelism, preserving diversity, and allowing adaptability with other search techniques.

Uploaded by

theoneandonlymandc
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
24 views2 pages

Genetic Algorithms - qn-2c

Genetic Algorithms (GA) are optimization tools inspired by natural evolution, utilizing mechanisms like population, chromosomes, genes, fitness functions, selection, crossover, and mutation to efficiently explore complex solution spaces. They enhance local search methods by simultaneously evaluating multiple solutions, maintaining diversity, and iterating through generations to improve overall fitness. GAs offer advantages over traditional local search by increasing parallelism, preserving diversity, and allowing adaptability with other search techniques.

Uploaded by

theoneandonlymandc
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

Genetic Algorithms(GA) are a powerful optimization tool inspired by the process of natural

evolution as they are particularly useful for exploring large and complex solution space
efficiently.
When adopted as a form of local search, GAs can leverage their unique mechanisms to improve
the search process.
Key Concepts of Genetic Algorithms
i. Population: A set of candidate solution
ii. Chromosomes: Representation of candidate solution, in form of strings or arrays
iii. Genes: components of chromosomes, representing variables or part of the solution
iv. Fitness Function: this is the function that evaluates and assign a fitness score to each
candidate solution based on how close it is to optimal solution
v. Selection: process of choosing filter candidate for reproduction
vi. Crossover(Recombination): combining parts of two parent solution to create
offspring
vii. Mutation: Randomly altering parts of a solution to introduce diversity

Local Search with Genetic Algorithms


Local search methods typically focus on iteratively improving a single solution by exploring its
neighborhood. Genetic Algorithms can be viewed as an extension of local search because they
simultaneously explore multiple solutions and their neighborhoods.
Here is how GAs efficiently explore solution space:
- Diverse Initial Population:
This involves initialization, where starts with diverse set of potential solutions, this
diversity ensures a wide coverage of the solution space thus preventing premature
convergence to suboptimal solutions.

- Fitness-Based Selection:
Selection Mechanism, which involves techniques like roulette wheel selection,
tournament selection, or rank-based favor fitter solutions. This ensures that better
solution is more likely to be carried forward thus improving the overall quality of the
population over time.

- Crossover(Exploration):
By combining parts of two parent solutions, crossover introduces new solutions that are
not present in the initial solution, this helps to explore new areas of the solution space
that may not be reachable by local moves from a single solution.

- Mutation (Exploitation and Exploration):


Random Alteration, through mutation it helps maintain genetic diversity within the
population. This allows the algorithm to explore new solutions and escape local optima
by making small, random changes to candidate solutions.

- Iteration and Convergence:


The process of selection, crossover and Mutation is repeated over many generations.
Each iteration improves the populations overall fitness, gradually converging towards
optimal or near-optimal solutions.
Advantages Over Local Traditional Local Search
- Parallelism: Genetic Algorithms work with a population of solutions simultaneously,
which increases the chances of finding a global optimum
- Diversity Preservation: with the use of techniques like mutation and crossover ensure a
diverse set of solutions, reducing the risk of getting stuck in local optima
- Adaptability: Genetic Algorithms with other local search techniques for example hill
climbing, simulated annealing can further enhance performance, leveraging the strength
of both global and local search methods.

You might also like