Scribd
Upload
2 views

Genetic Algorithms[1]

Uploaded by

xagey58277
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
2 views

Genetic Algorithms[1]

Uploaded by

xagey58277
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 15

Genetic Algorithms

Understanding GA Concepts, Working,


and Applications

By
58 - Janhavi Pawar
57 - Saraswati Shelke
51 - Het Patel
59 - Atharv Navle

Under Guidance Of
Prof. Jayshree Aher
Introduction
• Search heuristics inspired by natural selection.
• Used for optimization and search problems in complex domains.
• Simulates evolution to refine solutions over generations.
• Helps discover optimal or near-optimal solutions efficiently.
Why Use Genetic Algorithms?

• Suitable for large search spaces


• Can avoid local minima (unlike gradient
descent)
• Works well for nonlinear, complex problems
• Adaptable to various fields (ML, engineering,
etc.)
Genetic Algorithms –
Key Principles:-
• Inspired by Natural Selection – Mimics the evolutionary process found in
biology.
• Survival of the Fittest – Best solutions are selected for reproduction.
• Combination of Solutions – Selected solutions undergo crossover to create
new ones.
• Variation Introduction – Mutation introduces small changes to maintain
diversity.
• Iterative Improvement – Over multiple generations, solutions evolve toward
optimality.
Working of Genetic Algorithm

1. Initialize a random population


2. Evaluate fitness of each solution
3. Select best individuals for reproduction
4. Apply crossover & mutation
5. Repeat until convergence or stopping criteria
met
Selection Methods
• Roulette Wheel Selection – Probability-based
• Tournament Selection – Competes in small groups
• Rank Selection – Prioritizes higher-ranked solutions
• Elitism – Preserves best individuals

Key Components of GA
•Population – Set of potential solutions
•Fitness Function – Evaluates solution quality
•Selection – Chooses best solutions for reproduction
•Crossover – Combines solutions to create offspring
•Mutation – Introduces small random changes
Crossover Techniques
•Single-Point Crossover – A single cut is made in the parent
chromosomes, and the segments are swapped to create offspring.

•Multi-Point Crossover – Multiple cut points are chosen, and alternating


segments are exchanged between parents.

•Uniform Crossover – Each gene in the offspring is randomly inherited


from either parent with equal probability, ensuring high genetic diversity.

•Arithmetic Crossover – A mathematical combination of parent genes


(e.g., averaging or weighted sum) is used to create offspring, commonly
applied in numerical optimization problems.
Mutation Techniques in Genetic Algorithms

Bit Flip Mutation Swap Mutation

Gaussian Mutation Adaptive Mutation


Types of Genetic Algorithms (GAs)
1.Standard Genetic Algorithm (SGA)
2.Steady-State Genetic Algorithm (SSGA)
3.Micro-Genetic Algorithm (μGA)
4.Genetic Programming (GP)
5.Elitist Genetic Algorithm
6.Differential Evolution (DE)
7.Multi-Objective Genetic Algorithm (MOGA)
8.Cooperative Coevolutionary Genetic Algorithm
(CCGA)
9.Hybrid Genetic Algorithms
Multi-Objective Genetic Algorithm (MOGA)
• MOGA is designed to solve problems that have
multiple conflicting objectives.
• It aims to find a set of optimal solutions rather than just a
single solution.
•Key Advantages:

• Finds a set of diverse optimal solutions rather than just one.


• Ideal for problems where improving one objective can worsen
another (e.g., cost vs. quality).
How Multi-Objective Genetic Algorithm
(MOGA) Works
1.Initialize Population – Generate an initial population of potential solutions.
2.Evaluate Fitness – Evaluate solutions based on multiple conflicting objectives.
3.Pareto Dominance – Rank solutions using Pareto dominance to find non-dominated
solutions.
4.Selection – Select individuals for reproduction based on Pareto rank and diversity.
5.Crossover & Mutation – Apply crossover and mutation to create new solutions.
6.Replacement – Replace old population with new individuals while maintaining diversity.
7.Repeat – Continue the process until a stopping criterion (e.g., max generations or
convergence) is met.
Mathematical concepts related to MOGA:
1. Pareto Dominance

2. Pareto Optimality

3. Fitness Assignment (Pareto Front)

4. Crowding distance

5. Selection Based on Pareto Front


Select individuals based on Pareto rank and crowding distance
Advantages of Genetic Algorithms
•Global Optimization – Avoids local optima
•Flexibility – Works in diverse fields
•Parallelism – Can run multiple solutions
•No Derivatives Required – Useful for complex functions

Challenges & Limitations

•High Computational Cost – Needs multiple


iterations
•Parameter Sensitivity – Requires fine-tuning
•No Guarantee of Best Solution – Can get stuck in
suboptimal solutions
Applications of Genetic Algorithms
•Machine Learning – Feature selection, neural network
optimization
•Engineering – Structural design, control systems
•Bioinformatics – DNA sequence analysis
•Finance – Portfolio optimization
•Game Development – AI-driven strategies

Real-World Example
•Example: GA in traveling salesman problem (TSP)
•Optimizes shortest route for multiple cities
•Uses selection, crossover, and mutation to improve paths
Thank you!
CREDITS: This presentation template was created
by Slidesgo, and includes, infographics & images by
Freepik

You might also like