Introduction to Evolutionary Computation

1
Introduction to Evolutionary Computation

• Cheng-Hsiung Chiang, Assistant ProfessorDepartmen
  t of Information ManagementHsuan Chuang
  University
• Class E-mail chiang_at_imd.cc
• Class Homepage www.hcu.edu.tw/chchiang/Course/co
  urse.htm

2
Outlines

• What is evolutionary computation?
• Brief history of EC
• Taxonomy of EC
• Basic function of EC
• Applications of EC
• Advantages and disadvantages of EC

3
What is evolutionary computation?

• Evolutionary computation simulates evolution on a
  computer. The result of such a simulation is a
  series of optimization algorithms, usually based
  on a simple set of characteristics the
  equivalent of genome
• Recall that optimization iteratively improves the
  quality of solutions to some problem until an
  optimal (or at least feasible) solution is found
• Evolutionary Computation is the field of study
  devoted to the design, development, and analysis
  is problem solvers based on natural selection
  (simulated evolution).

4
A Brief Procedure of EC

• Procedure EC
• t 0
• Initialize Pop(t)
• Evaluate Pop(t)
• While (Not Done)
• Parents(t) Select_Parents(Pop(t))
• Offspring(t) Operating(Parents(t))
• Evaluate(Offspring(t))
• Pop(t1) Replace(Pop(t),Offspring(t))
• t t 1

CrossoverMutation, etc.
5
A Brief History of EC? Four Types

• L. Fogel 1962 (San Diego, CA) Evolutionary
  Programming
• J. Holland 1962 (Ann Arbor, MI)Genetic
  Algorithms
• I. Rechenberg H.-P. Schwefel 1965 (Berlin,
  Germany) Evolution Strategies
• J. Koza 1989 (Palo Alto, CA)Genetic Programming

6
Genetic Algorithms Genetic Programming

• Genetic algorithms (USA, 70s, Holland, DeJong)
• Typically applied to discrete optimization
• Attributed features
• not too fast
• good solver for combinatorial problems
• Special many variants, e.g., reproduction
  models, operators
• Genetic programming (USA, 90s, Koza)
• Typically applied to machine learning tasks
• Attributed features
• competes with neural nets and alike
• slow
• needs huge populations (thousands)
• Special non-linear chromosomes trees, graphs

7
Evolution Strategies Evolutionary Programming

• Evolution strategies (Germany, 70s, Rechenberg,
  Schwefel)
• Typically applied to numerical optimization
• Attributed features
• fast good optimizer for real-valued
  optimization
• relatively much theory
• Special self-adaptation of (mutation) parameters
  standard
• Evolutionary programming (USA, 60s, Fogel et
  al.)
• Typically applied to machine learning (old EP),
  optimization
• Attributed features very open framework any
  representation and mutation ops OK
• Special
• no recombination
• self-adaptation of parameters standard
  (contemporary EP)

8
The Generations of EC

• First Generation EC
• EP (Fogel)
• GA (Holland)
• ES (Rechenberg, Schwefel)
• EP (Fogel et al.)
• Second Generation EC
• Genetic Evolution of Data Structures
  (Michalewicz)
• Genetic Evolution of Programs (Koza)
• Hybrid Genetic Search (Davis)
• Tabu Search (Glover)

9
The Generations of EC

• Third Generation EC
• Artificial Immune Systems (Forrest)
• Cultural Algorithms (Reynolds)
• DNA Computing (Adleman)
• Ant Colony Optimization (Dorigo)
• Particle Swarm Optimization (Kennedy Eberhart)
• Memetic Algorithms
• Estimation of Distribution Algorithms
• Fourth Generation ????

10
Taxonomy of EC
Belongs to Artificial Intelligence
11
Basic function of EC
12
Components Representation / individuals (1)

• Individuals have two levels of existence
• phenotype object in original problem context,
  the outside
• genotype code to denote that object, the inside

phenotype
genotype
The link between these levels is called
representation
13
Components Representation / individiuals (2)
Genotype space
Phenotype space
Encoding (representation)
R 0 c 0 1 c d
B 0 c 0 1 c d
G 0 c 0 1 c d
Decoding (inverse representation)
14
Components selection

• Role
• Gives better individuals a higher chance of
• becoming parents
• surviving
• Pushes population towards higher fitness

E.g. roulette wheel selection
15
Components Mutation
Role causes small (random) variance
16
Components Recombination
Role combines features from different sources
17
Applications of EC

• Aircraft Design
• Routing in Communications Networks,
• Tracking Windshear,
• Game Playing
• Robotics
• Air Traffic Control
• Design
• Scheduling
• Machine Learning
• Pattern Recognition

• Market Forecasting
• Egg Price Forecasting
• Design of Filters
• Barriers
• Data-Mining
• User-Mining
• Resource Allocation
• Path Planning
• Job Shop Scheduling
• VLSI Circuit Layout
• Strike Force Allocation

18
Advantages of EC

• No presumptions w.r.t. problem space without
  math. model
• Widely applicable
• Low development application costs
• Easy to incorporate other methods
• Solutions are interpretable (unlike NN)
• Can be run interactively, accommodate user
  proposed solutions
• Provide many alternative solutions

19
Disadvantages of EC

• No guarantee for optimal solution within finite
  time
• Weak theoretical basis
• May need parameter tuning

20
????? ???