how should the GA proceed?

Size: px
Start display at page:

Download "how should the GA proceed?"

Transcription

1 how should the GA proceed? string fitness which new string would be better than any of the above? (the GA does not know the mapping between strings and fitness values!) recombine 'surface' similarities among the fittest strings combine 'good ideas' from different good strings because certain substructures in good strings cause their high fitness, and recombining such 'good ideas' may lead to better strings nb: fitness measure only says what needs to be done, not how to do it!! 1

2 how GAs work: population - variability - differential effect on survival find best strategy for running 4 hamburger joints strategy:! price: 50 or 5$? drink: wine or coke? service: slow or fast? restaurant 1: price=hi drink=coke service=fast 011 restaurant 2: hi wine fast 001 restaurant 3: low coke slow 110 restaurant 4: hi coke slow 010 unknown: which vbl most important? max profit? max loss? gradient info about fitness landscape (which single vbl, changed alone, has most impact)? can vbls be optimized separately or are they nonlinearly related? does environment itself change? available feedback: only weekly performance; no training period! since nothing is known about environment, test different random strategy in each restaurant for 1. week (hope you're not unlucky!). Favoring diversity expect payoff to approximate average payoff over whole search space 2

3 from generation 0 to generation 1 i string x i fitness f(x i ) f(x i ) / f(x i ) pool after reproduction 1 011! 3! f = total worst average! 1! 3.0!!!!!!! best!! 6!!!! 6 fitness-proportional reproduction: low-fit inds eliminated; expect ind 3 to be selected with prob.5 for each of the 4 positions in population, i.e. expect it to be selected for reproduction twice better average, better worst, less diversity crossing over strings 1 and 3: 01 1 xo 11 0: children 010 & 111. generation 1: 111 (7), 010 (2), 110 (6), 010 (2).. better best 3

4 fitness-proportional reproduction, xo, mut prob (select i ) = f i / f i string 1: 0.14 string 2: 0.49 string 3: 0.06 string 4: % 14% 6% 49% spin biased roulette wheel once for each string; copies of selected strings go into mating pool crossover: sexual reproduction mutation: 4

5 generic GA initialize population with random candidate solutions; evaluate fitness of each individual; while termination condition not satisfied do 1. select parents; 2. recombine pairs of parents; 3. mutate resulting offspring; 4. evaluate fitness of offspring; 5. select individuals for next generation; 1. fitness proportional, stochastic 2. whether to crossover, which parts, how to combine parts: stochastic. All higher organisms use xover; basis of breeding guarantees that genotype space is connected (eventual convergence) 5. based on fitness and/or age; sometimes deterministic: rank & select best n. 5

6 global EA, local HC memetic... individuals are improved by hill climbing during fitness assessment. * Lamarckian or Baldwin Effect would be better than Memetic. 1: t number of iterations to Hill-Climb 2: P Build Initial Population 3: Best 4: repeat 5: AssessFitness(P) 6: for each individual P i P do 7: P i Hill-Climb(P i ) for t iterations Replace P i in P 8: if Best = or Fitness(P i ) > Fitness(Best) then 9: Best P i 10: P Join(P, Breed(P)) 11: until Best is the ideal solution or we have run out of time 12: return Best 6

7 initializing population enforce diversity: make sure all individuals in initial population are unique -- when a new individual is made, check if it's already there, if so, throw it away, make another one of course not by O(n 2 ) scanning through entire population but: create hash table store individuals as keys in hash table, with arbitrary values (O(n)) if individual is not in hash table, hash it and add to population, else throw away 7

8 from one generation to the next j := 1; while j popsize do select mate1 from oldpop; select mate2 from oldpop; crossover (mate1, mate2); /*also mutate*/ child1 and child2 go into newpop; maintain # of xovers, parentage, other statistics; fitness child1 := objfunction(decode(child1)); fitness child2 := objfunction(decode(child2)); j := j + 2; /*assuming even popsize*/. generational replacement: oldpop := newpop. 8

9 no run = N? yes end run := 0 gen := 0 make initial random pop run++ termination sat for run? no i := 0 measure individual fitness no i := 0 yes i++ i = M? yes designate result for run Generic Generational Genetic Algorithm select genetic operation Pr Px Pm select one individual based on fitness select two individuals based on fitness select one individual based on fitness gen++ yes perform reproduction perform crossover perform mutation no i = M? i++ insert copy in new pop insert offspring in new pop insert mutant in new pop i++ pick 1 of the 3+ genetic operations (Pr+Px+Pm=1), then select 1 or 2 individuals based on fitness; reselection is allowed. 9

10 generic GA Algorithm 20 The Genetic Algorithm (GA) 1: popsize desired population size 2: P {} 3: for popsize times do 4: P P {new random individual} 5: Best 6: repeat 7: for each individual P i P do 8: AssessFitness(P i ) 9: if Best = or Fitness(P i ) > Fitness(Best) then 10: Best P i 11: Q {} Here s 12: for popsize/2 times do 13: Parent P a SelectWithReplacement(P) 14: Parent P b SelectWithReplacement(P) 15: Children C a, C b Crossover(Copy(P a ), Copy(P b )) 16: Q P {Mutate(C a ), Mutate(C b )} 17: P Q 18: until Best is the ideal solution or we have run out of time 19: return Best 10

11 making random bit strings and vectors 1: v a new vector v 1, v 2,...v l 2: for i from 1 to l do 3: if 0.5 > a random number chosen uniformly between 0.0 and 1.0 inclusive then 4: v i true 5: else 6: v i false 7: return v 1: min minimum desired vector element value 2: max maximum desired vector element value 3: v a new vector v 1, v 2,...v l 4: for i from 1 to l do 5: v i random number chosen uniformly between min and max inclusive 6: return v 11

12 mutating bit strings and vectors 1: p probability of flipping a bit Often p is set to 1/l 2: v boolean vector v 1, v 2,...v l to be mutated 3: for i from 1 to l do 4: if p random number chosen uniformly from 0.0 to 1.0 inclusive then 5: v i (v i ) 6: return v 1: v vector v 1, v 2,...v l to be convolved 2: p probability of adding noise to an element in the vector 3: r half-range of uniform noise 4: min minimum desired vector element value 5: max maximum desired vector element value 6: for i from 1 to l do 7: if p random number chosen uniformly from 0.0 to 1.0 then 8: repeat 9: n random number chosen uniformly from r to r inclusive 10: until min v i + n max 11: v i v i + n 12: return v p often = 1 12

13 1 and 2 point crossover 1: v first vector v 1, v 2,...v L to be crossed over 2: w second vector w 1, w 2,...w L to be crossed over 3: c random integer chosen uniformly from 1 to L inclusive 4: for i from c to L do 5: Swap the values of v i and w i 6: return v and w if c=1 or c = L, no xover happens 1: v first vector v 1, v 2,...v l to be crossed over 2: w second vector w 1, w 2,...w l to be crossed over 3: c random integer chosen uniformly from 1 to l inclusive 4: d random integer chosen uniformly from 1 to l inclusive 5: if c > d then 6: Swap c and d 7: for i from c to d 1 do 8: Swap the values of v i and w i 9: return v and w 13

14 integers to binary... in binary, different bits have different significance; so Hamming distance between consecutive integers is often 1. e.g. chance of changing 7 (0111) to 8 (1000) chance of changing 7 to 6 (0110). thus Gray coding: int[] bintogray (int[] b) { g[1] = b[1]; for (k=2; k m; k++) {g[k] = xor (b[k-1], b[k]);} return g; } integer standard Gray int[] graytobin (int[] g) { b[1] = value = g[1]; for (k=2; k m; k++) {if (g[k] == 1) value =!value; b[k]) = value;} return b; }

15 selection decreasing selection pressure: truncation tournament, ranking (size 2 tournament = linear ranking) fitness proportional uniform 2 rounds of selection: 1. who gets to mate? 2. who gets to survive? one of these rounds should use uniform selection or selection pressure is too high - high selection pressure greedy. 15

16 roulette wheel selection chromosome v i has fitness f i o population fitness F = k=1 popsize f k o selection probability p i for chromosome v i = f i / F o cumulative probability q i for chromosome v i =!! j=1 i p j, i=1,,popsize roulette wheel selection:! 1. r := random number from [0, 1]! 2. if r q 1, select first chromosome v 1 ; otherwise, select the! k th chromosome v k (2 k popsize) such that q k-1 < r q k 16

17 roulette wheel selection, cont. assume the population is sorted by fitness from 1 to popsize μ; calculate [q 1,..., q μ ] where q i = j=1 i p j ; thus q μ = 1.0. current_member := 1; while (current_member μ) do r := random value uniformly from [0,1]; i := 1; while (q i < r) do i := i + 1; mating_pool[current_member] := population[i]; current_member++; fitness proportional selection, esp. roulette wheel, is highly susceptible to fitness function transposition (e.g. adding a constant to all fitness values).

18 roulette wheel selection 1: perform once per generation 2: global p population copied into a vector of individuals p 1, p 2,..., p l 3: global f f1, f 2,..., f l fitnesses of individuals in p in the same order as p 4: if f is all 0.0s then Deal with all 0 fitnesses gracefully 5: Convert f to all 1.0s 6: for i from 2 to l do Convert f to a CDF. This will also cause fl = s, the sum of fitnesses. 7: f i f i + f i 1 8: perform each time 9: n random number from 0 to f l inclusive 10: for i from 2 to l do This could be done more efficiently with binary search 11: if f i 1 < n f i then 12: return p i 13: return p 1 CDF: cumulative distribution function 18

19 roulette wheel selection and 1-point xover as a linear (slow!) search through the roulette wheel with fitness-weighted slots: partialsum := 0.0; j := 0; /*population index*/ wheelposition := random * total fitness F; repeat j++; partialsum += fitness j until partialsum wheelposition or j = popsize; return j /*index of selected individual*/ select one individual cossover parent1, parent2; return child1, child2 (and mutate at the same time): if random p xover then pick xpoint between 1 & length -1 else xpoint := length; for j:= 1 to xpoint do child1[j] := mutate(parent1[j] with p mut ); child2[j] := mutate(parent2[j] with p mut ); if xpoint length then for j:= xpoint + 1 to length do child1[j] := mutate(parent2[j] with p mut ); child2[j] := mutate(parent1[j] with p mut ); 19

20 stochastic universal sampling SUS: another form of fitness-proportional selection SUS spins the wheel only once, with popsize µ equally spaced pointers: r = random number in [0, 1]; for (sum = i = 0; i < µ; i++) for (sum += ExpVal(i, t); sum > r; r++) select(i); ExpVal(i, t): expected # of times individual i is selected to reproduce, i.e. fitness(i) / average pop fitness. Each i reproduces at least ExpVal(i, t) times and at most ExpVal(i, t) times. drawbacks of fitness-proportional selection schemes (SUS, roulette wheel) - early, large fitness differences stifle exploration: premature convergence - later, when many individuals are similar, evolution slows down: random search - need 2 passes through population: 1. mean fitness ; 2. ExpVal 20

21 stochastic universal sampling 1: perform once per n individuals produced Usually n = l, that is, once per generation 2: global p copy of vector of individuals (our population) p 1, p 2,..., p l, shuffled randomly To shuffle a vector randomly, see Algorithm 26 3: global f f1, f 2,..., f l fitnesses of individuals in p in the same order as p 4: global index 0 5: if f is all 0.0s then 6: Convert f to all 1.0s 7: for i from 2 to l do Convert f to a CDF. This will also cause fl = s, the sum of fitnesses. 8: f i f i + f i 1 9: global value random number from 0 to f l /n inclusive 10: perform each time 11: while f index < value do 12: index index : value value + f l /n 14: return p index SUS is O(n), not O(n lg n) to select n individuals; more importantly, unlike fitness-proportional selection, it guarantees that very fit individuals (over s/n) are chosen at least once! 21

Chapter 8: Introduction to Evolutionary Computation

Chapter 8: Introduction to Evolutionary Computation Computational Intelligence: Second Edition Contents Some Theories about Evolution Evolution is an optimization process: the aim is to improve the ability of an organism to survive in dynamically changing

More information

Stochastic Search: Part 2. Genetic Algorithms. Vincent A. Cicirello. Robotics Institute. Carnegie Mellon University

Stochastic Search: Part 2. Genetic Algorithms. Vincent A. Cicirello. Robotics Institute. Carnegie Mellon University Stochastic Search: Part 2 Genetic Algorithms Vincent A. Cicirello Robotics Institute Carnegie Mellon University 5000 Forbes Avenue Pittsburgh, PA 15213 cicirello@ri.cmu.edu 1 The Genetic Algorithm (GA)

More information

Lecture 9 Evolutionary Computation: Genetic algorithms

Lecture 9 Evolutionary Computation: Genetic algorithms Lecture 9 Evolutionary Computation: Genetic algorithms Introduction, or can evolution be intelligent? Simulation of natural evolution Genetic algorithms Case study: maintenance scheduling with genetic

More information

Parallel Genetic Algorithms

Parallel Genetic Algorithms Parallel Genetic Algorithms for the Calibration of Financial Models Riccardo Gismondi June 13, 2008 High Performance Computing in Finance and Insurance Research Institute for Computational Methods Vienna

More information

Fundamentals of Genetic Algorithms

Fundamentals of Genetic Algorithms Fundamentals of Genetic Algorithms : AI Course Lecture 39 40, notes, slides www.myreaders.info/, RC Chakraborty, e-mail rcchak@gmail.com, June 01, 2010 www.myreaders.info/html/artificial_intelligence.html

More information

Crossover Techniques in GAs

Crossover Techniques in GAs Crossover Techniques in GAs Debasis Samanta Indian Institute of Technology Kharagpur dsamanta@iitkgp.ac.in 16.03.2018 Debasis Samanta (IIT Kharagpur) Soft Computing Applications 16.03.2018 1 / 1 Important

More information

Scaling Up. So far, we have considered methods that systematically explore the full search space, possibly using principled pruning (A* etc.).

Scaling Up. So far, we have considered methods that systematically explore the full search space, possibly using principled pruning (A* etc.). Local Search Scaling Up So far, we have considered methods that systematically explore the full search space, possibly using principled pruning (A* etc.). The current best such algorithms (RBFS / SMA*)

More information

CSC 4510 Machine Learning

CSC 4510 Machine Learning 10: Gene(c Algorithms CSC 4510 Machine Learning Dr. Mary Angela Papalaskari Department of CompuBng Sciences Villanova University Course website: www.csc.villanova.edu/~map/4510/ Slides of this presenta(on

More information

Lecture 15: Genetic Algorithms

Lecture 15: Genetic Algorithms Lecture 15: Genetic Algorithms Dr Roman V Belavkin BIS3226 Contents 1 Combinatorial Problems 1 2 Natural Selection 2 3 Genetic Algorithms 3 31 Individuals and Population 3 32 Fitness Functions 3 33 Encoding

More information

Search. Search is a key component of intelligent problem solving. Get closer to the goal if time is not enough

Search. Search is a key component of intelligent problem solving. Get closer to the goal if time is not enough Search Search is a key component of intelligent problem solving Search can be used to Find a desired goal if time allows Get closer to the goal if time is not enough section 11 page 1 The size of the search

More information

GENETIC ALGORITHM FOR CELL DESIGN UNDER SINGLE AND MULTIPLE PERIODS

GENETIC ALGORITHM FOR CELL DESIGN UNDER SINGLE AND MULTIPLE PERIODS GENETIC ALGORITHM FOR CELL DESIGN UNDER SINGLE AND MULTIPLE PERIODS A genetic algorithm is a random search technique for global optimisation in a complex search space. It was originally inspired by an

More information

Genetic Algorithms & Modeling

Genetic Algorithms & Modeling Genetic Algorithms & Modeling : Soft Computing Course Lecture 37 40, notes, slides www.myreaders.info/, RC Chakraborty, e-mail rcchak@gmail.com, Aug. 10, 2010 http://www.myreaders.info/html/soft_computing.html

More information

Genetic Learning of Firms in a Competitive Market

Genetic Learning of Firms in a Competitive Market FH-Kiel, Germany University of Applied Sciences Andreas Thiemer, e-mail: andreas.thiemer@fh-kiel.de Genetic Learning of Firms in a Competitive Market Summary: This worksheet deals with the adaptive learning

More information

Genetic Algorithms: Basic Principles and Applications

Genetic Algorithms: Basic Principles and Applications Genetic Algorithms: Basic Principles and Applications C. A. MURTHY MACHINE INTELLIGENCE UNIT INDIAN STATISTICAL INSTITUTE 203, B.T.ROAD KOLKATA-700108 e-mail: murthy@isical.ac.in Genetic algorithms (GAs)

More information

Evolutionary computation

Evolutionary computation Evolutionary computation Andrea Roli andrea.roli@unibo.it DEIS Alma Mater Studiorum Università di Bologna Evolutionary computation p. 1 Evolutionary Computation Evolutionary computation p. 2 Evolutionary

More information

Local Search & Optimization

Local Search & Optimization Local Search & Optimization CE417: Introduction to Artificial Intelligence Sharif University of Technology Spring 2017 Soleymani Artificial Intelligence: A Modern Approach, 3 rd Edition, Chapter 4 Outline

More information

Evolutionary Computation

Evolutionary Computation Evolutionary Computation - Computational procedures patterned after biological evolution. - Search procedure that probabilistically applies search operators to set of points in the search space. - Lamarck

More information

Genetic Algorithms and Genetic Programming Lecture 17

Genetic Algorithms and Genetic Programming Lecture 17 Genetic Algorithms and Genetic Programming Lecture 17 Gillian Hayes 28th November 2006 Selection Revisited 1 Selection and Selection Pressure The Killer Instinct Memetic Algorithms Selection and Schemas

More information

Matt Heavner CSE710 Fall 2009

Matt Heavner CSE710 Fall 2009 Matt Heavner mheavner@buffalo.edu CSE710 Fall 2009 Problem Statement: Given a set of cities and corresponding locations, what is the shortest closed circuit that visits all cities without loops?? Fitness

More information

Evolutionary Computation. DEIS-Cesena Alma Mater Studiorum Università di Bologna Cesena (Italia)

Evolutionary Computation. DEIS-Cesena Alma Mater Studiorum Università di Bologna Cesena (Italia) Evolutionary Computation DEIS-Cesena Alma Mater Studiorum Università di Bologna Cesena (Italia) andrea.roli@unibo.it Evolutionary Computation Inspiring principle: theory of natural selection Species face

More information

Evolutionary computation

Evolutionary computation Evolutionary computation Andrea Roli andrea.roli@unibo.it Dept. of Computer Science and Engineering (DISI) Campus of Cesena Alma Mater Studiorum Università di Bologna Outline 1 Basic principles 2 Genetic

More information

Data Warehousing & Data Mining

Data Warehousing & Data Mining 13. Meta-Algorithms for Classification Data Warehousing & Data Mining Wolf-Tilo Balke Silviu Homoceanu Institut für Informationssysteme Technische Universität Braunschweig http://www.ifis.cs.tu-bs.de 13.

More information

Evolutionary Functional Link Interval Type-2 Fuzzy Neural System for Exchange Rate Prediction

Evolutionary Functional Link Interval Type-2 Fuzzy Neural System for Exchange Rate Prediction Evolutionary Functional Link Interval Type-2 Fuzzy Neural System for Exchange Rate Prediction 3. Introduction Currency exchange rate is an important element in international finance. It is one of the chaotic,

More information

Incremental Stochastic Gradient Descent

Incremental Stochastic Gradient Descent Incremental Stochastic Gradient Descent Batch mode : gradient descent w=w - η E D [w] over the entire data D E D [w]=1/2σ d (t d -o d ) 2 Incremental mode: gradient descent w=w - η E d [w] over individual

More information

Genetic Algorithms. Donald Richards Penn State University

Genetic Algorithms. Donald Richards Penn State University Genetic Algorithms Donald Richards Penn State University Easy problem: Find the point which maximizes f(x, y) = [16 x(1 x)y(1 y)] 2, x, y [0,1] z (16*x*y*(1-x)*(1-y))**2 0.829 0.663 0.497 0.331 0.166 1

More information

V. Evolutionary Computing. Read Flake, ch. 20. Genetic Algorithms. Part 5A: Genetic Algorithms 4/10/17. A. Genetic Algorithms

V. Evolutionary Computing. Read Flake, ch. 20. Genetic Algorithms. Part 5A: Genetic Algorithms 4/10/17. A. Genetic Algorithms V. Evolutionary Computing A. Genetic Algorithms 4/10/17 1 Read Flake, ch. 20 4/10/17 2 Genetic Algorithms Developed by John Holland in 60s Did not become popular until late 80s A simplified model of genetics

More information

Genetic Algorithms. Seth Bacon. 4/25/2005 Seth Bacon 1

Genetic Algorithms. Seth Bacon. 4/25/2005 Seth Bacon 1 Genetic Algorithms Seth Bacon 4/25/2005 Seth Bacon 1 What are Genetic Algorithms Search algorithm based on selection and genetics Manipulate a population of candidate solutions to find a good solution

More information

Computational statistics

Computational statistics Computational statistics Combinatorial optimization Thierry Denœux February 2017 Thierry Denœux Computational statistics February 2017 1 / 37 Combinatorial optimization Assume we seek the maximum of f

More information

V. Evolutionary Computing. Read Flake, ch. 20. Assumptions. Genetic Algorithms. Fitness-Biased Selection. Outline of Simplified GA

V. Evolutionary Computing. Read Flake, ch. 20. Assumptions. Genetic Algorithms. Fitness-Biased Selection. Outline of Simplified GA Part 5A: Genetic Algorithms V. Evolutionary Computing A. Genetic Algorithms Read Flake, ch. 20 1 2 Genetic Algorithms Developed by John Holland in 60s Did not become popular until late 80s A simplified

More information

Local Search & Optimization

Local Search & Optimization Local Search & Optimization CE417: Introduction to Artificial Intelligence Sharif University of Technology Spring 2018 Soleymani Artificial Intelligence: A Modern Approach, 3 rd Edition, Chapter 4 Some

More information

[Read Chapter 9] [Exercises 9.1, 9.2, 9.3, 9.4]

[Read Chapter 9] [Exercises 9.1, 9.2, 9.3, 9.4] 1 EVOLUTIONARY ALGORITHMS [Read Chapter 9] [Exercises 9.1, 9.2, 9.3, 9.4] Evolutionary computation Prototypical GA An example: GABIL Schema theorem Genetic Programming Individual learning and population

More information

IV. Evolutionary Computing. Read Flake, ch. 20. Assumptions. Genetic Algorithms. Fitness-Biased Selection. Outline of Simplified GA

IV. Evolutionary Computing. Read Flake, ch. 20. Assumptions. Genetic Algorithms. Fitness-Biased Selection. Outline of Simplified GA IV. Evolutionary Computing A. Genetic Algorithms Read Flake, ch. 20 2014/2/26 1 2014/2/26 2 Genetic Algorithms Developed by John Holland in 60s Did not become popular until late 80s A simplified model

More information

Interplanetary Trajectory Optimization using a Genetic Algorithm

Interplanetary Trajectory Optimization using a Genetic Algorithm Interplanetary Trajectory Optimization using a Genetic Algorithm Abby Weeks Aerospace Engineering Dept Pennsylvania State University State College, PA 16801 Abstract Minimizing the cost of a space mission

More information

Optimal Operation of Large Power System by GA Method

Optimal Operation of Large Power System by GA Method Journal of Emerging Trends in Engineering and Applied Sciences (JETEAS) (1): 1-7 Scholarlink Research Institute Journals, 01 (ISSN: 11-7016) jeteas.scholarlinkresearch.org Journal of Emerging Trends in

More information

Intelligens Számítási Módszerek Genetikus algoritmusok, gradiens mentes optimálási módszerek

Intelligens Számítási Módszerek Genetikus algoritmusok, gradiens mentes optimálási módszerek Intelligens Számítási Módszerek Genetikus algoritmusok, gradiens mentes optimálási módszerek 2005/2006. tanév, II. félév Dr. Kovács Szilveszter E-mail: szkovacs@iit.uni-miskolc.hu Informatikai Intézet

More information

Local Search and Optimization

Local Search and Optimization Local Search and Optimization Outline Local search techniques and optimization Hill-climbing Gradient methods Simulated annealing Genetic algorithms Issues with local search Local search and optimization

More information

Biology 11 UNIT 1: EVOLUTION LESSON 2: HOW EVOLUTION?? (MICRO-EVOLUTION AND POPULATIONS)

Biology 11 UNIT 1: EVOLUTION LESSON 2: HOW EVOLUTION?? (MICRO-EVOLUTION AND POPULATIONS) Biology 11 UNIT 1: EVOLUTION LESSON 2: HOW EVOLUTION?? (MICRO-EVOLUTION AND POPULATIONS) Objectives: By the end of the lesson you should be able to: Describe the 2 types of evolution Describe the 5 ways

More information

Gender Separation and Mating Constraints

Gender Separation and Mating Constraints Gender Separation and Mating Constraints Dana Vrajitoru danav@cs.iusb.edu Intelligent Systems Laboratory, Computer and Information Sciences Department, Indiana University South Bend, 1700 Mishawaka Ave,

More information

The Story So Far... The central problem of this course: Smartness( X ) arg max X. Possibly with some constraints on X.

The Story So Far... The central problem of this course: Smartness( X ) arg max X. Possibly with some constraints on X. Heuristic Search The Story So Far... The central problem of this course: arg max X Smartness( X ) Possibly with some constraints on X. (Alternatively: arg min Stupidness(X ) ) X Properties of Smartness(X)

More information

Development. biologically-inspired computing. lecture 16. Informatics luis rocha x x x. Syntactic Operations. biologically Inspired computing

Development. biologically-inspired computing. lecture 16. Informatics luis rocha x x x. Syntactic Operations. biologically Inspired computing lecture 16 -inspired S S2 n p!!! 1 S Syntactic Operations al Code:N Development x x x 1 2 n p S Sections I485/H400 course outlook Assignments: 35% Students will complete 4/5 assignments based on algorithms

More information

A.I.: Beyond Classical Search

A.I.: Beyond Classical Search A.I.: Beyond Classical Search Random Sampling Trivial Algorithms Generate a state randomly Random Walk Randomly pick a neighbor of the current state Both algorithms asymptotically complete. Overview Previously

More information

The Role of Crossover in Genetic Algorithms to Solve Optimization of a Function Problem Falih Hassan

The Role of Crossover in Genetic Algorithms to Solve Optimization of a Function Problem Falih Hassan The Role of Crossover in Genetic Algorithms to Solve Optimization of a Function Problem Falih Hassan ABSTRACT The genetic algorithm is an adaptive search method that has the ability for a smart search

More information

Schema Theory. David White. Wesleyan University. November 30, 2009

Schema Theory. David White. Wesleyan University. November 30, 2009 November 30, 2009 Building Block Hypothesis Recall Royal Roads problem from September 21 class Definition Building blocks are short groups of alleles that tend to endow chromosomes with higher fitness

More information

Evolutionary Computation: introduction

Evolutionary Computation: introduction Evolutionary Computation: introduction Dirk Thierens Universiteit Utrecht The Netherlands Dirk Thierens (Universiteit Utrecht) EC Introduction 1 / 42 What? Evolutionary Computation Evolutionary Computation

More information

Lecture 22. Introduction to Genetic Algorithms

Lecture 22. Introduction to Genetic Algorithms Lecture 22 Introduction to Genetic Algorithms Thursday 14 November 2002 William H. Hsu, KSU http://www.kddresearch.org http://www.cis.ksu.edu/~bhsu Readings: Sections 9.1-9.4, Mitchell Chapter 1, Sections

More information

Unit 1A: Computational Complexity

Unit 1A: Computational Complexity Unit 1A: Computational Complexity Course contents: Computational complexity NP-completeness Algorithmic Paradigms Readings Chapters 3, 4, and 5 Unit 1A 1 O: Upper Bounding Function Def: f(n)= O(g(n)) if

More information

Bounded Approximation Algorithms

Bounded Approximation Algorithms Bounded Approximation Algorithms Sometimes we can handle NP problems with polynomial time algorithms which are guaranteed to return a solution within some specific bound of the optimal solution within

More information

Mechanisms of Evolution

Mechanisms of Evolution Mechanisms of Evolution 36-149 The Tree of Life Christopher R. Genovese Department of Statistics 132H Baker Hall x8-7836 http://www.stat.cmu.edu/ ~ genovese/. Plan 1. Two More Generations 2. The Hardy-Weinberg

More information

Multi-objective genetic algorithm

Multi-objective genetic algorithm Multi-objective genetic algorithm Robin Devooght 31 March 2010 Abstract Real world problems often present multiple, frequently conflicting, objectives. The research for optimal solutions of multi-objective

More information

Binary Particle Swarm Optimization with Crossover Operation for Discrete Optimization

Binary Particle Swarm Optimization with Crossover Operation for Discrete Optimization Binary Particle Swarm Optimization with Crossover Operation for Discrete Optimization Deepak Singh Raipur Institute of Technology Raipur, India Vikas Singh ABV- Indian Institute of Information Technology

More information

Genetic Algorithm. Outline

Genetic Algorithm. Outline Genetic Algorithm 056: 166 Production Systems Shital Shah SPRING 2004 Outline Genetic Algorithm (GA) Applications Search space Step-by-step GA Mechanism Examples GA performance Other GA examples 1 Genetic

More information

Individual learning and population evolution

Individual learning and population evolution Genetic Algorithms [Read Chapter 9] [Exercises 9.1, 9.2, 9.3, 9.4] Evolutionary computation Prototypical GA An example: GABIL Genetic Programming Individual learning and population evolution 168 lecture

More information

Artificial Intelligence (AI) Common AI Methods. Training. Signals to Perceptrons. Artificial Neural Networks (ANN) Artificial Intelligence

Artificial Intelligence (AI) Common AI Methods. Training. Signals to Perceptrons. Artificial Neural Networks (ANN) Artificial Intelligence Artificial Intelligence (AI) Artificial Intelligence AI is an attempt to reproduce intelligent reasoning using machines * * H. M. Cartwright, Applications of Artificial Intelligence in Chemistry, 1993,

More information

Genetic Algorithm for Solving the Economic Load Dispatch

Genetic Algorithm for Solving the Economic Load Dispatch International Journal of Electronic and Electrical Engineering. ISSN 0974-2174, Volume 7, Number 5 (2014), pp. 523-528 International Research Publication House http://www.irphouse.com Genetic Algorithm

More information

I N N O V A T I O N L E C T U R E S (I N N O l E C) Petr Kuzmič, Ph.D. BioKin, Ltd. WATERTOWN, MASSACHUSETTS, U.S.A.

I N N O V A T I O N L E C T U R E S (I N N O l E C) Petr Kuzmič, Ph.D. BioKin, Ltd. WATERTOWN, MASSACHUSETTS, U.S.A. I N N O V A T I O N L E C T U R E S (I N N O l E C) Binding and Kinetics for Experimental Biologists Lecture 2 Evolutionary Computing: Initial Estimate Problem Petr Kuzmič, Ph.D. BioKin, Ltd. WATERTOWN,

More information

CSC242: Intro to AI. Lecture 5. Tuesday, February 26, 13

CSC242: Intro to AI. Lecture 5. Tuesday, February 26, 13 CSC242: Intro to AI Lecture 5 CSUG Tutoring: bit.ly/csug-tutoring League of Legends LAN Party: Sat 2/2 @ 2PM in CSB 209 $2 to benefit Big Brothers Big Sisters bit.ly/urlol ULW Topics due to me by tomorrow

More information

CHAPTER 4 INTRODUCTION TO DISCRETE VARIABLE OPTIMIZATION

CHAPTER 4 INTRODUCTION TO DISCRETE VARIABLE OPTIMIZATION CHAPTER 4 INTRODUCTION TO DISCRETE VARIABLE OPTIMIZATION. Introduction.. Examples of Discrete Variables One often encounters problems in which design variables must be selected from among a set of discrete

More information

Representation and Hidden Bias II: Eliminating Defining Length Bias in Genetic Search via Shuffle Crossover

Representation and Hidden Bias II: Eliminating Defining Length Bias in Genetic Search via Shuffle Crossover Representation and Hidden Bias II: Eliminating Defining Length Bias in Genetic Search via Shuffle Crossover Abstract The traditional crossover operator used in genetic search exhibits a position-dependent

More information

LandscapeEC: Adding Geographical Structure to Cellular Evolutionary Algorithms

LandscapeEC: Adding Geographical Structure to Cellular Evolutionary Algorithms LandscapeEC: Adding Geographical Structure to Cellular Evolutionary Algorithms Lucas Ellgren and Nicholas Freitag McPhee Division of Science and Mathematics University of Minnesota, Morris Morris, MN 562367

More information

Local Beam Search. CS 331: Artificial Intelligence Local Search II. Local Beam Search Example. Local Beam Search Example. Local Beam Search Example

Local Beam Search. CS 331: Artificial Intelligence Local Search II. Local Beam Search Example. Local Beam Search Example. Local Beam Search Example 1 S 331: rtificial Intelligence Local Search II 1 Local eam Search Travelling Salesman Problem 2 Keeps track of k states rather than just 1. k=2 in this example. Start with k randomly generated states.

More information

Evolutionary Design I

Evolutionary Design I Evolutionary Design I Jason Noble jasonn@comp.leeds.ac.uk Biosystems group, School of Computing Evolutionary Design I p.1/29 This lecture Harnessing evolution in a computer program How to construct a genetic

More information

Evolutionary Algorithms: Introduction. Department of Cybernetics, CTU Prague.

Evolutionary Algorithms: Introduction. Department of Cybernetics, CTU Prague. Evolutionary Algorithms: duction Jiří Kubaĺık Department of Cybernetics, CTU Prague http://cw.felk.cvut.cz/doku.php/courses/a4m33bia/start pcontents 1. duction to Evolutionary Algorithms (EAs) Pioneers

More information

S0214 : GPU Based Stacking Sequence Generation For Composite Skins Using GA

S0214 : GPU Based Stacking Sequence Generation For Composite Skins Using GA S0214 : GPU Based Stacking Sequence Generation For Composite Skins Using GA Date: 16th May 2012 Wed, 3pm to 3.25pm(Adv. Session) Sathyanarayana K., Manish Banga, and Ravi Kumar G. V. V. Engineering Services,

More information

Geometric Semantic Genetic Programming (GSGP): theory-laden design of variation operators

Geometric Semantic Genetic Programming (GSGP): theory-laden design of variation operators Geometric Semantic Genetic Programming (GSGP): theory-laden design of variation operators Andrea Mambrini University of Birmingham, UK NICaiA Exchange Programme LaMDA group, Nanjing University, China 7th

More information

School of EECS Washington State University. Artificial Intelligence

School of EECS Washington State University. Artificial Intelligence School of EECS Washington State University Artificial Intelligence 1 } Focused on finding a goal state Less focused on solution path or cost } Choose a state and search nearby (local) states Not a systematic

More information

Motivation. Dictionaries. Direct Addressing. CSE 680 Prof. Roger Crawfis

Motivation. Dictionaries. Direct Addressing. CSE 680 Prof. Roger Crawfis Motivation Introduction to Algorithms Hash Tables CSE 680 Prof. Roger Crawfis Arrays provide an indirect way to access a set. Many times we need an association between two sets, or a set of keys and associated

More information

REIHE COMPUTATIONAL INTELLIGENCE COLLABORATIVE RESEARCH CENTER 531

REIHE COMPUTATIONAL INTELLIGENCE COLLABORATIVE RESEARCH CENTER 531 U N I V E R S I T Y OF D O R T M U N D REIHE COMPUTATIONAL INTELLIGENCE COLLABORATIVE RESEARCH CENTER 531 Design and Management of Complex Technical Processes and Systems by means of Computational Intelligence

More information

CSC 421: Algorithm Design & Analysis. Spring 2018

CSC 421: Algorithm Design & Analysis. Spring 2018 CSC 421: Algorithm Design & Analysis Spring 2018 Complexity & Computability complexity theory tractability, decidability P vs. NP, Turing machines NP-complete, reductions approximation algorithms, genetic

More information

Estimation-of-Distribution Algorithms. Discrete Domain.

Estimation-of-Distribution Algorithms. Discrete Domain. Estimation-of-Distribution Algorithms. Discrete Domain. Petr Pošík Introduction to EDAs 2 Genetic Algorithms and Epistasis.....................................................................................

More information

A GENETIC ALGORITHM FOR FINITE STATE AUTOMATA

A GENETIC ALGORITHM FOR FINITE STATE AUTOMATA A GENETIC ALGORITHM FOR FINITE STATE AUTOMATA Aviral Takkar Computer Engineering Department, Delhi Technological University( Formerly Delhi College of Engineering), Shahbad Daulatpur, Main Bawana Road,

More information

Evolutionary Algorithms: Introduction. The Czech Institute of Informatics, Robotics and Cybernetics CTU Prague

Evolutionary Algorithms: Introduction. The Czech Institute of Informatics, Robotics and Cybernetics CTU Prague : Introduction Jiří Kubaĺık The Czech Institute of Informatics, Robotics and Cybernetics CTU Prague http://cw.felk.cvut.cz/doku.php/courses/a0m33eoa/start pcontents Representation Operators Population

More information

Using Evolutionary Techniques to Hunt for Snakes and Coils

Using Evolutionary Techniques to Hunt for Snakes and Coils Using Evolutionary Techniques to Hunt for Snakes and Coils Abstract The snake-in-the-box problem is a difficult problem in mathematics and computer science that deals with finding the longest-possible

More information

mrna Codon Table Mutant Dinosaur Name: Period:

mrna Codon Table Mutant Dinosaur Name: Period: Mutant Dinosaur Name: Period: Intro Your dinosaur is born with a new genetic mutation. Your job is to map out the genes that are influenced by the mutation and to discover how the new dinosaurs interact

More information

DETECTING THE FAULT FROM SPECTROGRAMS BY USING GENETIC ALGORITHM TECHNIQUES

DETECTING THE FAULT FROM SPECTROGRAMS BY USING GENETIC ALGORITHM TECHNIQUES DETECTING THE FAULT FROM SPECTROGRAMS BY USING GENETIC ALGORITHM TECHNIQUES Amin A. E. 1, El-Geheni A. S. 2, and El-Hawary I. A **. El-Beali R. A. 3 1 Mansoura University, Textile Department 2 Prof. Dr.

More information

CS 473: Algorithms. Ruta Mehta. Spring University of Illinois, Urbana-Champaign. Ruta (UIUC) CS473 1 Spring / 32

CS 473: Algorithms. Ruta Mehta. Spring University of Illinois, Urbana-Champaign. Ruta (UIUC) CS473 1 Spring / 32 CS 473: Algorithms Ruta Mehta University of Illinois, Urbana-Champaign Spring 2018 Ruta (UIUC) CS473 1 Spring 2018 1 / 32 CS 473: Algorithms, Spring 2018 Universal Hashing Lecture 10 Feb 15, 2018 Most

More information

CS 331: Artificial Intelligence Local Search 1. Tough real-world problems

CS 331: Artificial Intelligence Local Search 1. Tough real-world problems S 331: rtificial Intelligence Local Search 1 1 Tough real-world problems Suppose you had to solve VLSI layout problems (minimize distance between components, unused space, etc.) Or schedule airlines Or

More information

CSC242: Artificial Intelligence. Lecture 4 Local Search

CSC242: Artificial Intelligence. Lecture 4 Local Search CSC242: Artificial Intelligence Lecture 4 Local Search Upper Level Writing Topics due to me by next class! First draft due Mar 4 Goal: final paper 15 pages +/- 2 pages 12 pt font, 1.5 line spacing Get

More information

Reinforcement Learning: Part 3 Evolution

Reinforcement Learning: Part 3 Evolution 1 Reinforcement Learning: Part 3 Evolution Chris Watkins Department of Computer Science Royal Holloway, University of London July 27, 2015 2 Cross-entropy method for TSP Simple genetic style methods can

More information

The European Applied Business Research Conference Rothenburg, Germany 2002

The European Applied Business Research Conference Rothenburg, Germany 2002 The European Applied Business Research Conference Rothenburg, Germany 00 Using Genetic Algorithms To Optimise Kanban-Based Production System M Al-Tahat, (E-mail: mohammadal-tahat@mailinguniboit), University

More information

Population Genetics & Evolution

Population Genetics & Evolution The Theory of Evolution Mechanisms of Evolution Notes Pt. 4 Population Genetics & Evolution IMPORTANT TO REMEMBER: Populations, not individuals, evolve. Population = a group of individuals of the same

More information

Cell division and multiplication

Cell division and multiplication CELL DIVISION Cell division and multiplication As we already mentioned, the genetic information contained in the nucleus is hereditary Meaning it is passed on from cell to cell; from parent to child This

More information

Chapter 13 Meiosis and Sexual Reproduction

Chapter 13 Meiosis and Sexual Reproduction Biology 110 Sec. 11 J. Greg Doheny Chapter 13 Meiosis and Sexual Reproduction Quiz Questions: 1. What word do you use to describe a chromosome or gene allele that we inherit from our Mother? From our Father?

More information

CSE 202 Homework 4 Matthias Springer, A

CSE 202 Homework 4 Matthias Springer, A CSE 202 Homework 4 Matthias Springer, A99500782 1 Problem 2 Basic Idea PERFECT ASSEMBLY N P: a permutation P of s i S is a certificate that can be checked in polynomial time by ensuring that P = S, and

More information

Grade 9 District Formative Assessment-Extended Response. Name Teacher

Grade 9 District Formative Assessment-Extended Response. Name Teacher Name Teacher /5 ER.DFA1.9.RST.04 Determine the meaning of symbols, key terms, and other domain-specific words and phrases as they are used in a specific scientific or technical context relevant to grades

More information

Landscapes and Other Art Forms.

Landscapes and Other Art Forms. Landscapes and Other Art Forms. Darrell Whitley Computer Science, Colorado State University Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted

More information

Generalization of Dominance Relation-Based Replacement Rules for Memetic EMO Algorithms

Generalization of Dominance Relation-Based Replacement Rules for Memetic EMO Algorithms Generalization of Dominance Relation-Based Replacement Rules for Memetic EMO Algorithms Tadahiko Murata 1, Shiori Kaige 2, and Hisao Ishibuchi 2 1 Department of Informatics, Kansai University 2-1-1 Ryozenji-cho,

More information

Local search algorithms

Local search algorithms Local search algorithms CS171, Winter 2018 Introduction to Artificial Intelligence Prof. Richard Lathrop Reading: R&N 4.1-4.2 Local search algorithms In many optimization problems, the path to the goal

More information

Computational Intelligence Winter Term 2018/19

Computational Intelligence Winter Term 2018/19 Computational Intelligence Winter Term 2018/19 Prof. Dr. Günter Rudolph Lehrstuhl für Algorithm Engineering (LS 11) Fakultät für Informatik TU Dortmund Three tasks: 1. Choice of an appropriate problem

More information

Evolutionary Algorithms

Evolutionary Algorithms Evolutionary Algorithms a short introduction Giuseppe Narzisi Courant Institute of Mathematical Sciences New York University 31 January 2008 Outline 1 Evolution 2 Evolutionary Computation 3 Evolutionary

More information

Evolving a New Feature for a Working Program

Evolving a New Feature for a Working Program Evolving a New Feature for a Working Program Mike Stimpson arxiv:1104.0283v1 [cs.ne] 2 Apr 2011 January 18, 2013 Abstract A genetic programming system is created. A first fitness function f 1 is used to

More information

Reproduction- passing genetic information to the next generation

Reproduction- passing genetic information to the next generation 166 166 Essential Question: How has biological evolution led to the diversity of life? B-5 Natural Selection Traits that make an organism more or less likely to survive in an environment and reproduce

More information

Haploid-Diploid Algorithms

Haploid-Diploid Algorithms Haploid-Diploid Algorithms Larry Bull Department of Computer Science & Creative Technologies University of the West of England Bristol BS16 1QY, U.K. +44 (0)117 3283161 Larry.Bull@uwe.ac.uk LETTER Abstract

More information

Local and Stochastic Search

Local and Stochastic Search RN, Chapter 4.3 4.4; 7.6 Local and Stochastic Search Some material based on D Lin, B Selman 1 Search Overview Introduction to Search Blind Search Techniques Heuristic Search Techniques Constraint Satisfaction

More information

Haploid & diploid recombination and their evolutionary impact

Haploid & diploid recombination and their evolutionary impact Haploid & diploid recombination and their evolutionary impact W. Garrett Mitchener College of Charleston Mathematics Department MitchenerG@cofc.edu http://mitchenerg.people.cofc.edu Introduction The basis

More information

Towards Automatic Design of Adaptive Evolutionary Algorithms. Ayman Srour - Patrick De Causmaecker

Towards Automatic Design of Adaptive Evolutionary Algorithms. Ayman Srour - Patrick De Causmaecker Towards Automatic Design of Adaptive Evolutionary Algorithms Ayman Srour - Patrick De Causmaecker Outline Background Parameter tuning Adaptive Parameter control Preliminary investigation The proposed framework

More information

Metaheuristics and Local Search

Metaheuristics and Local Search Metaheuristics and Local Search 8000 Discrete optimization problems Variables x 1,..., x n. Variable domains D 1,..., D n, with D j Z. Constraints C 1,..., C m, with C i D 1 D n. Objective function f :

More information

How robust are the predictions of the W-F Model?

How robust are the predictions of the W-F Model? How robust are the predictions of the W-F Model? As simplistic as the Wright-Fisher model may be, it accurately describes the behavior of many other models incorporating additional complexity. Many population

More information

www.lessonplansinc.com Topic: Dinosaur Evolution Project Summary: Students pretend to evolve two dinosaurs using genetics and watch how the dinosaurs adapt to an environmental change. This is a very comprehensive

More information

Introduction to Evolutionary Computation

Introduction to Evolutionary Computation Introduction to Evolutionary Computation 1 Evolutionary Computation Different lines of investigation in simulated evolution GA (Genetic Algorithms) ES (Evolution Strategies) EP (Evolutionary Programming)

More information

Performance Evaluation of Best-Worst Selection Criteria for Genetic Algorithm

Performance Evaluation of Best-Worst Selection Criteria for Genetic Algorithm Mathematics and Computer Science 2017; 2(6): 89-97 http://www.sciencepublishinggroup.com/j/mcs doi: 10.11648/j.mcs.20170206.12 ISSN: 2575-6036 (Print); ISSN: 2575-6028 (Online) Performance Evaluation of

More information