Evolutionary Computation

1. Evolutionary Computation Artificial Intelligence Department of Industrial Engineering and Management Cheng Shiu University

2. Outline • Can evolution be intelligent? • Simulation of natural evolution • Natural Genetics • Evolutionary Cycle • Developments of Evolutionary computation • Ideas of Evolutionary methods • Essential Elements of EC

3. Can evolution be intelligent? • Intelligence can be defined as the capability of a system to adapt its behaviour to ever-changing environment. According to Alan Turing, the form or appearance of a system is irrelevant to its intelligence. • Evolutionary computation simulates evolution on a computer. The result of such a simulation is a series of optimisation algorithms, usually based on a simple set of rules. Optimisation iteratively improves the quality of solutions until an optimal, or at least feasible, solution is found.

4. The behaviour of an individual organism is an inductive inference about some yet unknown aspects of its environment. If, over successive generations, the organism survives, we can say that this organism is capable of learning to predict changes in its environment. • The evolutionary approach is based on computational models of natural selection and genetics. We call them evolutionary computation, an umbrella term that combines genetic algorithms,evolution strategiesandgenetic programming.

5. Simulation of natural evolution • On 1 July 1858, Charles Darwin presented his theory of evolution before the Linnean Society of London. This day marks the beginning of a revolution in biology. • Darwin’s classical theory of evolution, together with Weismann’s theory of natural selection and Mendel’s concept of genetics, now represent the neo-Darwinian paradigm.

6. Neo-Darwinism is based on processes of reproduction, mutation, competition and selection. The power to reproduce appears to be an essential property of life. The power to mutate is also guaranteed in any living organism that reproduces itself in a continuously changing environment. Processes of competition and selection normally take place in the natural world, where expanding populations of different species are limited by a finite space.

7. Evolution can be seen as a process leading to the maintenance of a population’s ability to survive and reproduce in a specific environment. This ability is called evolutionary fitness. • Evolutionary fitness can also be viewed as a measure of the organism’s ability to anticipate changes in its environment. • The fitness, or the quantitative measure of the ability to predict environmental changes and respond adequately, can be considered as the quality that is optimised in natural life.

8. How is a population with increasing fitness generated? • Let us consider a population of rabbits. Some rabbits are faster than others, and we may say that these rabbits possess superior fitness, because they have a greater chance of avoiding foxes, surviving and then breeding. • If two parents have superior fitness, there is a good chance that a combination of their genes will produce an offspring with even higher fitness. Over time the entire population of rabbits becomes faster to meet their environmental challenges in the face of foxes.

9. Natural Genetics • The information required to build a living organism is coded in the DNA and other genetic material found in the cells of that organism • Within a species, most of the genetic material is the same • Small changes in the genetic material give rise to small changes in the organism • E.g height, hair colour

10. DNA and Genes • DNA is a large molecule made up of fragments. There are several fragment types, each one acting like a letter in a long coded message: -A-B-A-D-C-B-B-C-C-A-D-B-C-C-A- • Certain groups of letters are meaningful together - a bit like words. • These groups are called genes • The DNA is made up of genes and rubbish A,T,C,G

11. Example: Human Reproduction • Human DNA is organised into chromosomes • Most human cells contains 23 pairs of chromosomes which together define the physical attributes of the individual:

12. Reproductive Cells • Sperm and egg cells contain 23 individual chromosomes rather than 23 pairs • Reproductive cells are formed by one cell splitting into two • During this process the pairs of chromosome undergo an operation called crossover

13. Crossover During crossover the chromosome pairs link up and swap parts of themselves: Before After After crossover one of each pair goes into each cell

14. Fertilisation Sperm cell from Father Egg cell from Mother New person cell

15. Mutation • Occasionally some of the genetic material changes very slightly during this process • This means that the child might have genetic material information not inherited from either parent • This is most likely to be catastrophic

16. Parents Population Offspring The Evolutionary Cycle Selection Evaluation Recombination Mutation Replacement

17. Simulation of natural evolution • All methods of evolutionary computation simulate natural evolution by creating a population of individuals, evaluating their fitness, generating a new population through genetic operations, and repeating this process a number of times. • We will start with Genetic Algorithms (GAs) as most of the other evolutionary algorithms can be viewed as variations of genetic algorithms.

18. CI and Evolutionary Algorithms

19. 50‘s and 60‘s The Michigan School The German School The California School (Holland) (Rechenberg, Schwefel, Bienert) (Fogel, Owens, Walsh) - More emphasis on recombination, inversion, etc. - Used the term “genetics” - More emphasis on selection and mutation - Used the term “evolution” evolution strategies simulated evolution Genetic Algorithms Evolution Strategies Evolutionary Programming t o d a y Genetic Programming + (Koza et al.): Ideas of the Michigan School put to automated programming

20. History 1859 Charles Darwin: inheritance, variation, natural selection 1957 G. E. P. Box: random mutation & selection for optimization 1958 Fraser, Bremermann: computer simulation of evolution 1964 Rechenberg, Schwefel: mutation & selection 1966 Fogel et al.: evolving automata - “evolutionary programming” 1975 Holland: crossover, mutation & selection - “reproductive plan” 1975 De Jong: parameter optimization - “genetic algorithm” 1989 Goldberg: first textbook 1991 Davis: first handbook 1993 Koza: evolving LISP programs - “genetic programming”

21. Historical Perspective • Evolutionary computation • Invented 10 times independently from 1953 - 1970 • Early pioneers included Barricelli, Bledsoe, Bremermann, Box, Conrad, Fogel, Fraser, Friedberg, Holland, Kaufman, Rechenberg, Schwefel and others.

22. Evolutionary Computation Fraser(1957); Bremmerman (1958) Fogel (1962) Fogel et al (1966) Schwefel & Rechenberg (1965) Holland (1975) Schwefel (1981) Goldberg (1989) D. Fogel (1992) Koza(1992)

24. General Ideas of Evolutionary Methods • EA’s use nature as an inspiration • the principles of evolution • birth and death of individuals in a population • survival of the fittest • multiple generations • mating or ‘crossover’ (two parents creating one offspring) • mutation (random variation) programming

25. EVOLUTION Individual Fitness Environment PROBLEM SOLVING Candidate Solution Quality Problem The Evolutionary Metaphor

26. A Pseudo Program of Evolutionary Algorithms begin t : = 0 ; initialize P( t ) ; begin While not terminate do evaluate P( t ) ; begin M( t ) : = select-mates( P( t ) ) ; O( t ) : = alteration( M( t ) ) ; evaluate( O( t ) ) ; P( t+1 ) : = select( O( t )∪P( t ) ) ; t : = t+1 ; end end end

27. A Pseudo Program of Evolutionary Computation (2) • Evolutionary computation starts with an initialization of a population of individuals (solution candidates), called P(0), with a population size to be supplied by the users. • These solution will then be evaluated based an objective function or a fitness function determined by the problem we encounter.

28. A Pseudo Program of Evolutionary Computation (3) • The continuation of the procedure will hinges on the termination criteria supplied by users. • Two major operators are conducted to form the new generation, which can be as considered as a correspondence as follows:

29. Three Stages • First Stage: making a mating pool (M(t)) • Second Stage: generating offspring (O(t)) • Third Stage: forming new generation (P(t+1))

30. Pseudo code of an Evolutionary Algorithm begin EA; t = 0; // Initializing time // Initialize a usually random population of individuals: random P( t ); // Evaluate the fitness of all individuals: evaluate P( t ); whilenot done do// Testing for termination criterion t = t + 1; // Increasing time // Select a sub-population for offspring production: P' = select P( t ); // Recombine the “genes” of selected parents: recombine P'( t ); // Stochastically perturb genes of the mated population: mutate P'( t ); // Evaluate the new fitness: evaluate P'( t ); // Select the survivors from actual fitness: P = survive P( t ), P'( t ) endwhile; end EA;

31. Pseudo Code of an Evolutionary Algorithm Create initial random population Evaluate fitness of each individual yes Termination criteria satisfied ? stop no Select parents according to fitness Recombine parents to generate offspring Mutate offspring Replace population by new offspring

32. The Evolutionary Cycle initialize population evaluate select mating partners (terminate) recombinate loop select mutate evaluate

33. Selection Parents Recombination Population Mutation Replacement Offspring

34. Object problem and Fitness solution genotype s M f fitness object problem

35. Five Essential Elements of EC • individuals and their representations, • population size, • selection, • mutation, • recombination.

36. Five Essential Elements of EC mutation recombination t + 1 t reproduction selection