GENETIC ALGORITHMS - I
Dr. Priyanka Parimi
Dr. Pallavi Mishra
Contents
1. Optimization Problems
2. Concept of "Genetics" and "Evolution"
3. Basic GA framework
4. Different GA architectures
5. GA operators: Encoding, Crossover, Selection, Mutation
6. Solving single objective optimization problems using GAs
7. Case Studies and Applications
Optimization Problems
Optimization
Problem
The goal is to find a values of
x1, x2,…, xn which give the
maximum (or minimum) objective
under the given constraints
Benchmark Optimization Problems
➢Traveling Salesman Problem
➢Knapsack Problem
➢Graph Coloring Problem
➢Job Machine Assignment Problem
➢Coin Change Problem
➢Binary search tree construction problem
Travelling
Salesman
Problem
Shortest Path Problem
Types of Optimization Problems
➢Unconstrained optimization Problems
➢Constrained Optimization problems
➢Integer Programming problems
➢Real-valued problems
➢Mixed-integer programming problem
➢Linear optimization problems
➢Non-linear optimization problems
Discrete Optimization Problems
➢Minimum spanning tree problem ➢Network Design Problems
➢Shortest path problem ➢Facility Location Problems
➢Maximum flow problem ➢Inventory management
➢Min-cost flow problem ➢Transportation Problems
➢Assignment Problem
➢Computationally expensive.
Limitations of ➢For a discontinuous objective function, methods
Traditional may fail.
Optimization ➢Method may not be suitable for parallel
Approaches computing.
➢Discrete (integer) variables are difficult to handle.
➢Methods may not necessarily adaptive.
Soft Computing techniques have been evolved to
address the above mentioned limitations of solving
optimization problem with traditional approaches.
Evolutionary Algorithms
Algorithms inspired by Biologic behaviours:
➢Genetics and Evolution –> Genetic Algorithms (GA)
➢Behavior of ant colony –> Ant Colony Optimization (ACO)
➢Human nervous system –> Artificial Neural Network (ANN)
Algorithms inspired by Physical behaviours:
➢Annealing process –> Simulated Annealing (SA)
➢Swarming of particle –> Particle Swarming Optimization (PSO)
➢Learning –> Fuzzy Logic (FL)
Concept of "Genetics"
and "Evolution"
Background of Genetic Algorithm
Introduced by Prof. John Holland (of Michigan University, USA, 1965).
But, the first article on GA was published in 1975.
Principles of GA based on two fundamental biological processes:
Genetics: Gregor Johan Mendel (1865)
Evolution: Charles Darwin (1875)Debasis
Spiral helix of protein substance is called DNA.
For a specie, DNA code is unique, that is, vary uniquely from one
to other.
DNA code (inherits some characteristics from one generation to
Genetic Code next generation) is used as biometric trait.
Genetics
The basic building blocks in living bodies are cells.
Each cell carries the basic unit of heredity, called
gene
For a particular specie, number of chromosomes is
fixed.
Examples
◦ Mosquito: 6
◦ Frogs: 26
◦ Human: 46
◦ Goldfish: 94
Reproduction
Crossing-over
Four primary premises:
➢Information propagation: An offspring has many of its
characteristics of its parents (i.e. information passes
from parent to its offspring). [Heredity]
➢Population diversity: Variation in characteristics in the
Evolution next generation. [Diversity]
Evolution is a process Natural ➢Survival for existence: Only a small percentage of the
Selection offspring produced survive to adulthood. [Selection]
➢Survival of the best: Offspring survived depends on
their inherited characteristics. [Ranking]
Mutation:
To make the process forcefully dynamic when variations
in population going to stable.
Biological
Evolution
process
Genetic Algorithm (GA)
Genetic Algorithm
Genetic algorithm (GA) is a population-based probabilistic search oriented
optimization technique, which works based on the principles of Genetics and
Natural Selection.
It is frequently used to find optimal or near-optimal solutions to difficult
problems otherwise would take a lifetime to solve
GAs are a subset of much larger branch of computation known as Evolutionary
Computation
Genetic Algorithm
➢GA is an iterative process.
➢It is a searching technique.
Important ➢Working cycle with / without convergence.
Features of GA ➢Solution is not necessarily guaranteed.
➢Usually, terminated with a local optima.
➢Not very fast
Benefits of GA
➢Easy to Understand ➢Many ways to speed up and improce GA-
based application as knowledge about
➢Modular, separate from application problem domain is gained
➢Supports multi-objective optimization ➢Easy to exploit previous or alternate solutions
➢Good for “noisy” environments ➢Flexible building blocks for hybrid
➢Always an answer, answer gets better with applications
time ➢Substantial history and range of use
➢Inherently parallel and easily distributed
Basic framework for
Genetic Algorithm
Basic framework of Genetic Algorithm
Genetic Algorithm
Terminology
Population – It is a subset of all the
possible (encoded) solutions to the
given problem
Chromosomes – A chromosome is
one such solution to the given
problem
Gene – A gene is one element
position of a chromosome
Allele – It is the value a gene takes
for a particular chromosome
Terminology
➢Genotype – Genotype is the population in the computational space.
➢The solutions are represented in a way which can be easily undertood nad manipulated using a
computing system
➢Phenotype – phenotype is the population in the actual real world solution space.
➢The solutions are represented in a way they are shown in real world situations
➢For simple problems, genotype and phenotype are same.
➢Encoding is a process of transforming from the phenotype to genotype space,
➢Decoding is a process of transforming a solution from genotype to phenotype space.
➢Decoding should be fast as it is carried out repeatedly in a GA during fitness evaluation
Terminology
➢Fitness function is a function which takes the solution as input and produces the suitability of
the solution as the output.
➢In some cases, fitness function and the objective function may be the same, where as it may be
different in others based n the problem.
➢Genetic Operators are used to alter the genetic composition of the offspring.
➢These include selection, crossover, mutation, inversion, etc.
Detailed
Framework of
GA
GA Operations
Encoding: How to represent a solution to fit with GA framework.
Convergence: How to decide the termination criterion.
Mating pool: How to generate next solutions.
Fitness Evaluation: How to evaluate a solution.
Crossover: How to make the diverse set of next solutions.
Mutation: To explore other solution(s).
Inversion: To move from one optima to other.
GA architectures
Generational
Genetic
Algorithm (GGA)
Parameters for GGA
➢Initial population size : N
➢Size of mating pool, Np : Np = p% of N
➢Convergence threshold δ
➢Mutation μ
➢Inversion η
➢Crossover ρ
Characteristics of GGA
➢Have overlapping generation (Only fraction of individuals are replaced).
➢Computationally expensive.
➢Good when initial population size is large.
➢In general, gives better results.
➢Selection is biased toward more highly fit individuals; Hence, the average fitness (of overall population) is
expected to increase in succession.
➢The best individual may appear in any iteration.
Steady-State
Genetic
Algorithm
(SSGA)
Characteristics of SSGA
➢Generation gap is small.
➢Only two offspring are produced in one generation.
It is applicable when
➢Population size is small
➢Chromosomes are of longer length
➢Evaluation operation is less computationally expensive (compare to duplicate checking)
Limitations of SSGA
➢There is a chance of stuck at local optima, if crossover/ mutation/ inversion is not strong
enough to diversify the population.
➢Premature convergence may result.
➢It is susceptible to stagnation. Inferiors are neglected or removed and keeps making more trials
for very long period of time without any gain (i.e. long period of localized search).
