Genetic Algorithms & Evolutionary Computation Flashcards
Describe the steps of an evolutionary/genetic algorithm.
Initialize population, evaluate, select parents, do crossover and mutation, select survivors, terminate population.
How does the evolutionary computing metaphor map biological concepts to problem-solving concepts?
Evolution == Problem Solving,
Environment == Problem,
Individual == Candidate solution,
Fitness == Quality.
What is the distinction between genotype and phenotype in biological evolution?
Genotype is the encoded information, phenotype is the actual manifestation.
What is the role of mutation in genetic algorithms?
To create random variations to keep genetic diversity.
How does evolutionary computation handle problems where the majority of the search space is infeasible, such as university timetabling?
The algorithm discovers and refines feasible solutions by steering the reproduction and mutation toward constraint satisfaction.
What is the role of the fitness function in genetic algorithms?
It calculates how good an individual/chromosome/solution is, measuring how well it meets the objective/constraints.
How does swarm intelligence differ from other evolutionary computation approaches?
Swarm intelligence relies on teamwork and shared information between agents. Evolutionary computation relies on competition and survival of the fittest.
What advantages do evolutionary algorithms offer over traditional problem-solving methods?
Evolutionary algorithms are flexible and can find solutions in rather large infeasible search spaces, and doesn’t require problem-specific knowledge.
What role does population size and diversity play in evolutionary algorithms?
Population size affects the efficiency; larger populations allow more searched solutions. Diversity ensures more exploration of the solution space.
Explain how ‘survival of the fittest’ influenced the design of genetic algorithms.
Survival of the fittest influenced selection and fitness function. It mimics by selecting only better solutions for reproduction, ensuring advancement toward the solution.
Explain what a chromosome is in a genetic algorithm.
A chromosome represents a possible solution within the search space.
What is the role of crossover in genetic algorithms?
The role of crossover is to take the traits of two better solutions and mix them, creating an offspring that may be better.
What is evolution?
Evolution is the process of change in organisms over generations, to better adapt to their environment.
What is the role of selection in genetic algorithms?
Selection chooses which individuals to reproduce, favouring the better fitness individuals.
What is the elitism mechanism in genetic algorithms?
It prevents the loss of progress by ensuring some of the best individuals get passed on to the next generation.
Design a chromosome representation for the class scheduling problem from assignment
2. Explain each of the components of your representation. How would crossover and mutation work for your representation?
A chromosome would contain a list of genes. A gene contains 3 ints representing the indexes which correspond to each list of courses, rooms, and timeslots.
Ex. Gene = (0,26) and a chromosome is many genes creating a timetable that may be the soltution to the problem
