Introduction to genetic programming

Introduction to genetic programming

• 1.
  What is GeneticProgramming  Genetic programming is a systematic method for getting computers to automatically solve a problem starting from a high-level statement of what needs to be done.  Genetic programming iteratively transforms a population of computer programs into a new generation of programs by applying analogs of naturally occurring genetic operations.  The genetic operations include crossover (sexual recombination), mutation, reproduction.
• 2.
  What is GeneticProgramming  Programs are expressed in genetic programming as syntax trees rather than as lines of code. For example, the simple expression  max(x∗x,x + 3∗y)
• 3.
  What is GeneticProgramming  Tree include nodes and links. Node represent the instruction to execute and link represent arguments for each instruction.  In the following the internal nodes in a tree will be called functions, while the tree’s leaves will be called terminals.
• 4.
  Why Genetic Programming?  It saves time by freeing the human from having to design complex algorithms. Not only designing the algorithms but creating ones that give optimal solutions.  It combines genetic algorithms with the basic thrust of AI, which was to get computers to do things automatically – to evolve a population of programs.  It is difficult, unnatural, and overly restrictive to attempt to represent hierarchies of dynamically varying size and shape with fixed length character strings.  For many problems in machine learning and artificial intelligence, the most natural representation for a solution is a computer program.
• 5.
  PREPARATORY STEPS OFGENETIC PROGRAMMING  the set of terminals for each branch of the to-be-evolved program,  the set of primitive functions for each branch of the to-be-evolved program,  the fitness measure (for explicitly or implicitly measuring the fitness of individuals in the population),  certain parameters for controlling the run, and  the termination criterion and method for designating the result of the run.
• 6.
  STEPS To PerformGenetic Programming 1) Randomly create an initial population of individual computer programs composed of the available functions and terminals. 2) Iteratively perform the following sub-steps (called a generation) on the population - A:- Execute each program in the population and ascertain its fitness (explicitly or implicitly) using the problem’s fitness measure. B:- Select one or two individual program(s) from the population with a probability based on fitness to participate in the genetic operations.
• 7.
  STEPS To PerformGenetic Programming C:- Create new individual program(s) for the population by applying the following genetic operations- (I) Reproduction: Copy the selected individual program to the new population. (II) Crossover: Create new offspring program(s) for the new population by recombining randomly chosen parts from two selected programs. (III) Mutation: Create one new offspring program for the new population by randomly mutating a randomly chosen part of one selected program.
• 8.
  STEPS To PerformGenetic Programming  After the termination criterion is satisfied, the single best program in the population produced during the run (the best-so-far individual) is harvested and designated as the result of the run.
• 9.
   Genetic Operationsexample-
• 10.
  Differences Between GPand Genetic Algorithm  The main difference between genetic programming and genetic algorithms is the representation of the solution. Genetic programming creates computer programs in the lisp or scheme computer languages as the solution.  Genetic algorithms create a string of numbers that represent the solution. Genetic programming, uses four steps to solve problems.  Genetic programming is much more powerful than genetic algorithms. The output of the genetic algorithm is a quantity, while the output of the genetic programming is a another computer program.
• 11.
  Differences Between GPand Genetic Algorithm  Genetic programming works best for several types of problems.  The first type is where there is no ideal solution, (for example, a program that drives a car).  There is no one solution to driving a car. Some solutions drive safely at the expense of time, while others drive fast at a high safety risk. Therefore, driving a car consists of making compromises of speed versus safety, as well as many other variables.  In this case genetic programming will find a solution that attempts to compromise and be the most efficient solution from a large list of variables.
• 12.
  Real world Applications Neural Network Optimization [Zhang, Mühlenbein, 1993]  Image Analysis [Poli, 1996a]  Generation of a knowledge base for expert systems [Bojarczuk, Lopes, Freitas, 2000]  Fuzzy Logic Control [Akbarzadeh, Kumbla, Tunstel, Jamshidi, 2000]  Hardware Evolution (Field-Programmable Gate Array) [Thompson, 1997].  Symbolic Regression  Problem: Given a set of data points, find a mathematical model