Introduction To Genetic Algorithms

Transcription

1 1 Introduction To Genetic Algorithms Dr. Rajib Kumar Bhattacharjya Department of Civil Engineering IIT Guwahati

2 References 2 D. E. Goldberg, Genetic Algorithm In Search, Optimization And Machine Learning, New York: Addison Wesley (1989) John H. Holland Genetic Algorithms, Scientific American Journal, July Kalyanmoy Deb, An Introduction To Genetic Algorithms, Sadhana, Vol. 24 Parts 4 And 5.

3 Introduction to optimization 3 Global optima Local optima Local optima Local optima Local optima f X

4 Introduction to optimization 4 Multiple optimal solutions

5 Genetic Algorithms 5 Genetic Algorithms are the heuristic search and optimization techniques that mimic the process of natural evolution.

6 Principle Of Natural Selection 6 Select The Best, Discard The Rest

7 An Example. 7 Giraffes have long necks Giraffes with slightly longer necks could feed on leaves of higher branches when all lower ones had been eaten off. They had a better chance of survival. Favorable characteristic propagated through generations of giraffes. Now, evolved species has long necks.

8 An Example. 8 This longer necks may have due to the effect of mutation initially. However as it was favorable, this was propagated over the generations.

9 Evolution of species 9 Initial Population of animals Millions Of Years Struggle For Existence Survival Of the Fittest Surviving Individuals Reproduce, Propagate Favorable Characteristics Evolved Species

10 10 Thus genetic algorithms implement the optimization strategies by simulating evolution of species through natural selection

11 Simple Genetic Algorithms 11 Start Initialize population Evaluate Solutions Optimum Solution? T = 0 YES NO Selection T=T+1 Stop Crossover Mutation

12 Simple Genetic Algorithm 12 function sga () { Initialize population; Calculate fitness function; While(fitness value!= termination criteria) { Selection; Crossover; Mutation; } Calculate fitness function; }

13 GA Operators and Parameters 13 Selection Crossover Mutation Now we will discuss about genetic operators

14 Selection 14 The process that determines which solutions are to be preserved and allowed to reproduce and which ones deserve to die out. The primary objective of the selection operator is to emphasize the good solutions and eliminate the bad solutions in a population while keeping the population size constant. Selects the best, discards the rest

15 Functions of Selection operator 15 Identify the good solutions in a population Make multiple copies of the good solutions Eliminate bad solutions from the population so that multiple copies of good solutions can be placed in the population Now how to identify the good solutions?

16 Fitness function 16 A fitness value can be assigned to evaluate the solutions A fitness function value quantifies the optimality of a solution. The value is used to rank a particular solution against all the other solutions A fitness value is assigned to each solution depending on how close it is actually to the optimal solution of the problem

17 Assigning a fitness value 17 Considering c = h d

18 Selection operator 18 There are different techniques to implement selection in Genetic Algorithms. They are: Tournament selection Roulette wheel selection Proportionate selection Rank selection Steady state selection, etc

19 Tournament selection 19 In tournament selection several tournaments are played among a few individuals. The individuals are chosen at random from the population. The winner of each tournament is selected for next generation. Selection pressure can be adjusted by changing the tournament size. Weak individuals have a smaller chance to be selected if tournament size is large.

20 Tournament selection Selected Best solution will have two copies Worse solution will have no copies Other solutions will have two, one or zero copies

21 21 Roulette wheel and proportionate selection Parents are selected according to their fitness values The better chromosomes have more chances to be selected Chrom # Fitness Roulet wheel 16% % of RW % EC AC % 2 3% 4 18% 3 21%

22 Rank selection 22 Chrom # Fitness Sort according to fitness Chrom # Fitness Assign raking Chrom # Rank Chrom # % of RW Roulette wheel Chrom # EC AC

23 Steady state selection 23 In this method, a few good chromosomes are used for creating new offspring in every iteration. Then some bad chromosomes are removed and the new offspring is placed in their places The rest of population migrates to the next generation without going through the selection process. Good New offspring Good Bad New offspring

24 How to implement crossover 24 The crossover operator is used to create new solutions from the existing solutions available in the mating pool after applying selection operator. This operator exchanges the gene information between the solutions in the mating pool Encoding of solution is necessary so that our solutions look like a chromosome Source:

25 Encoding 25 The process of representing a solution in the form of a string that conveys the necessary information. Just as in a chromosome, each gene controls a particular characteristic of the individual, similarly, each bit in the string represents a characteristic of the solution.

26 Encoding Methods 26 Most common method of encoding is binary coded. Chromosomes are strings of 1 and 0 and each position in the chromosome represents a particular characteristic of the problem Decoded value Mapping between decimal and binary value

27 Encoding Methods 27 Defining a string [ ] d h h (d,h) = (8,10) cm Chromosome = [ ] d

28 Crossover operator 28 The most popular crossover selects any two solutions strings randomly from the mating pool and some portion of the strings is exchanged between the strings. Solution 1 The selection point is selected randomly. A probability of crossover is also introduced in order to give freedom to an individual solution string to determine whether the solution would go for crossover or not. Solution 2 Child 1 Child 2

29 Binary Crossover 29 Source: Deb 1999

30 Mutation operator 30 Mutation is the occasional introduction of new features in to the solution strings of the population pool to maintain diversity in the population. Though crossover has the main responsibility to search for the optimal solution, mutation is also used for this purpose. Before mutation After mutation

31 Binary Mutation 31 Mutation operator changes a 1 to 0 or vise versa, with a mutation probability of. The mutation probability is generally kept low for steady convergence. A high value of mutation probability would search here and there like a random search technique. Source: Deb 1999

32 Elitism 32 Crossover and mutation may destroy the best solution of the population pool Elitism is the preservation of few best solutions of the population pool Elitism is defined in percentage or in number

33 Nature to Computer Mapping 33 Nature Population Individual Fitness Chromosome Gene Computer Set of solutions Solution to a problem Quality of a solution Encoding for a solution Part of the encoding solution

34 An example problem 34 Consider 6 bit string to represent the solution, then = 0 and = Assume population size of 4 Let us solve this problem by hand calculation

35 An example problem 35 Initialize population Calculate decoded value Calculate real value Calculate objective function value Calculate actual count Calculate expected count Calculate relative fitness value Calculate fitness value Initial population Sol No Binary String Decoding DV x value Fitness calculation f F Selection: Proportionate selection Relative Fitness Expected count Actual count Avg Max

36 An example problem: Crossover 36 Matting pool Random generation of crossover site New population Crossover: Single point Sol No Matting pool CS New Binary String DV x value f F Avg Max 0.97

37 An example problem: Mutation 37 Mutation Sol No Population after crossover Population after mutation DV x value f F Avg Max 1.00

38 Real coded Genetic Algorithms 38 Disadvantage of binary coded GA more computation lower accuracy longer computing time solution space discontinuity hamming cliff

39 Real coded Genetic Algorithms 39 The standard genetic algorithms has the following steps 1. Choose initial population 2. Assign a fitness function 3. Perform elitism 4. Perform selection 5. Perform crossover 6. Perform mutation In case of standard Genetic Algorithms, steps 5 and 6 require bitwise manipulation.

40 Real coded Genetic Algorithms 40 Simple crossover: similar to binary crossover P1 P C1 C

41 Real coded Genetic Algorithms 41 Linear Crossover Parents: (x 1,,x n ) and (y 1,,y n ) Select a single gene (k) at random Three children are created as, ( x 1,..., x k, 0.5 y k x k,..., x n ) x,..., x,1.5 y 0.5 x,..., x ) ( 1 k k k n ( x 1,..., x k, y k x k,..., x n From the three children, best two are selected for the next generation )

42 Real coded Genetic Algorithms 42 Single arithmetic crossover Parents: (x 1,,x n ) and (y 1,,y n ) Select a single gene (k) at random child 1 is created as, ( x 1,..., x, α k y k + (1 α) x k,..., x n reverse for other child. e.g. with α = 0.5 )

43 Real coded Genetic Algorithms 43 Simple arithmetic crossover Parents: (x 1,,x n ) and (y 1,,y n ) Pick random gene (k) after this point mix values child 1 is created as: ( x,..., x, α y 1 k k + (1 α) + 1 x k,..., α + 1 reverse for other child. e.g. with α = 0.5 y n + (1 α) x n )

44 Real coded Genetic Algorithms 44 Whole arithmetic crossover Most commonly used Parents: (x 1,,x n ) and (y 1,,y n ) child 1 is: α x + ( 1 α) y reverse for other child. e.g. with α =

45 Simulated binary crossover 45 Developed by Deb and Agrawal, 1995) Where, a random number is a parameter that controls the crossover process. A high value of the parameter will create near-parent solution

46 Random mutation 46 Where is a random number between [0,1] Where, is the user defined maximum perturbation

47 Normally distributed mutation 47 A simple and popular method Where is the Gaussian probability distribution with zero mean

48 Polynomial mutation 48 Deb and Goyal,1996 proposed, =, + 2 / 1 = / If <0.5 If 0.5

49 Multi-modal optimization 49 Solve this problem using simple Genetic Algorithms

50 After Generation The population are in and around the global optimal solution

51 Multi-modal optimization 51 Simple modification of Simple Genetic Algorithms can capture all the optimal solution of the problem including global optimal solutions Basic idea is that reduce the fitness of crowded solution, which can be implemented using following three steps. Sharing function Niche count Modified fitness

52 Hand calculation 52 Maximize Sol String Decoded value x f

53 Distance table 53 dij

54 Sharing function values 54 sh(dij) nc

55 Sharing fitness value 55 Sol String Decoded value x f nc f

56 56 Solutions obtained using modified fitness value

57 Evolutionary Strategies 57 ES use real parameter value ES does not use crossover operator It is just like a real coded genetic algorithms with selection and mutation operators only

58 Two members ES: (1+1) ES 58 In each iteration one parent is used to create one offspring by using Gaussian mutation operator

59 Two members ES: (1+1) ES 59 Step1: Choose a initial solution and a mutation strength Step2: Create a mutate solution Step 3: If, replace with Step4: If termination criteria is satisfied, stop, else go to step 2

60 Two members ES: (1+1) ES 60 Strength of the algorithm is the proper value of Rechenberg postulate The ratio of successful mutations to all the mutations should be 1/5. If this ratio is greater than 1/5, increase mutation strength. If it is less than 1/5, decrease the mutation strength.

61 Two members ES: (1+1) ES 61 A mutation is defined as successful if the mutated offspring is better than the parent solution. If is the ratio of successful mutation over n trial, Schwefel (1981) suggested a factor in the following update rule

62 Matlab code 62

63 63

64 Some results of 1+1 ES Y Optimal Solution is X*= [ ] Objective function value f = X

65 Multimember ES 65 ES Step1: Choose an initial population of solutions and mutation strength Step2: Create mutated solution Step3: Combine and, and choose the best solutions Step4: Terminate? Else go to step 2

66 Multimember ES 66 ES Through selection Through mutation

67 Multimember ES 67 ES Offspring Through selection Through mutation

68 Multi-objective optimization 68 Comfort Price

69 Multi-objective optimization 69 Two objectives are Minimize weight Minimize deflection

70 Multi-objective optimization 70 More than one objectives Objectives are conflicting in nature Dealing with two search space Decision variable space Objective space Unique mapping between the objectives and often the mapping is non-linear Properties of the two search space are not similar Proximity of two solutions in one search space does not mean a proximity in other search space

71 Multi-objective optimization 71

72 Vector Evaluated Genetic Algorithm (VEGA) 72 f1 f2 f3 f4 P1 P2 P3 P4 Crossover and Mutation fn Pn Old population Mating pool New population Propose by Schaffer (1984)

73 Non-dominated selection heuristic 73 Give more emphasize on the non-dominated solutions of the population This can be implemented by subtracting from the dominated solution fitness value Suppose is the number of sub-population and is the non-dominated solutions. Then total reduction is. The total reduction is then redistributed among the non-dominated solution by adding an amount / This method has two main implications Non-dominated solutions are given more importance Additional equal emphasis has been given to all the non-dominated solution

74 Weighted based genetic algorithm (WBGA) 74 The fitness is calculated The spread is maintained using the sharing function approach Sharing function Niche count Modified fitness

75 Multiple objective genetic algorithm (MOGA) x2 2.5 f x1 Solution space f1 Objective space

76 Multiple objective genetic algorithm (MOGA) Maximize f2 Maximize f1

77 Multiple objective genetic algorithm (MOGA) 77 Fonseca and Fleming (1993) first introduced multiple objective genetic algorithm (MOGA) The assigned fitness value based on the non-dominated ranking. The rank is assigned as =1+ where is the ranking of the solution and is the number of solutions that dominate the solution

78 Multiple objective genetic algorithm (MOGA) 78 Fonseca and Fleming (1993) maintain the diversity among the non-dominated solution using niching among the solution of same rank. The normalize distance was calculated as, The niche count was calculated as,

79 NSGA 79 Srinivas and Deb (1994) proposed NSGA The algorithm is based on the non-dominated sorting. The spread on the Pareto optimal front is maintained using sharing function

80 NSGA II 80 Non-dominated Sorting Genetic Algorithms NSGA II is an elitist non-dominated sorting Genetic Algorithm to solve multi-objective optimization problem developed by Prof. K. Deb and his student at IIT Kanpur. It has been reported that NSGA II can converge to the global Pareto-optimal front and can maintain the diversity of population on the Pareto-optimal front

81 Non-dominated sorting 81 1 Objective 2 (Minimize) Objective 2 (Minimize) Feasible Region 3 Infeasible Region Objective 1 (Minimize) Objective 1 (Minimize)

82 Calculation crowding distance 82 Objective Cd, the crowded distance is the perimeter of the rectangle constituted by the two neighboring solutions Cd value more means that the solution is less crowded Objective 1 Cd value less means that the solution is more crowded

83 Crowded tournament operator 83 A solution wins a tournament with another solution, If the solution has better rank than, i.e. < If they have the same rank, but has a better crowding distance than, i.e. = and >.

84 Replacement scheme of NSGA II P Selected Q 84 Rejected Non dominated sorting Rejected based on crowding distance

85 Initialize population of size N Calculate all the objective functions 85 Rank the population according to nondominating criteria No Termination Criteria? Yes Pareto-optimal solution Selection Crossover Mutation Calculate objective function of the new population Combine old and new population Non-dominating ranking on the combined population Calculate crowding distance of all the solutions Get the N member from the combined population on the basis of rank and crowding distance Replace parent population by the better members of the combined population

86 Constraints handling in GA f(x) X ( )

87 Constraints handling in GA f(x) X ( ) +

88 Constraints handling in GA f(x) X + ( ) +

89 Constrain handling in GA 89 Minimize Subject to 0 h =0 =1,2,3,, =1,2,3,, Deb s approach = = + + h If is feasible Otherwise

90 90 THANKS