A survival of the fittest principle will naturally emerge given:
- A population of organisms which can reproduce in a challenging/changing environment
- A way of continually generating diversity in new “child” organisms
The basic method of evolution as a solution to a problem
Trial and error, but this appears to be a recipe for problem solving algorithms which take forever, with little or no eventual success.
(no heuristics or propagation or feedback loop)
Infinite Monkey Theorem
Given an infinite length of time, a chimpanzee punching at random on a typewriter would almost surely type out all of Shakespeare’s plays.
Although randomness (stochasticity) is involved in an evolutionary algorithm, other ingredients, inspired by nature are required to develop a functioning algorithm
Those 3 ingredients are:
1) Use a population of organisms that are competing for resources
2) Select “parents” with a relatively weak bias towards the fittest
3) It can sometimes help to use recombination of two or more “parents” - i.e. generate new candidate solutions by combining bits and pieces from different previous solutions.
1&2 are required. 3 is optional, but is often helpful
“Select ‘parents’ with a relatively weak bias towards the fittest.”
It’s not really plain survival of the fittest, what works is the fitter you are, the more chance you have to reproduce, and it works best if even the least fit still have some chance. Mutate these, i.e. apply a small change
The basic techniques used when adapting a gene pool through the process of evolution
Selection
Crossover
Mutation
These 3 techniques are applied iteratively to take an initial population of genes and create children, which are then the new population
Examples of evolutionary algorithms aiding in the bentley’s thesis on car design
Fixed wheel position, constrained bounding area
Chromosome is a series of slices
Fitnesses evaluated via a simple airflow simulation
Evolution Algorithm applications in planning
Routing
Scheduling
Packing
Evolution Algorithm applications in design
Electronic circuits
Neural networks
Structure design
Evolution Algorithm applications in simulations
Model economic interactions of competing firms in a market
Evolution Algorithm applications in identification
Fit a function medical data to predict future values
Evolution Algorithm applications in control
Design a controller for gas turbine engine
Design control system for mobile robots
Evolution Algorithm applications in classification
Game playing
Diagnosis of heart disease
Detecting SPAM
How Nature Inspired (Genetic and Evolutionary) algorithms differ from machine learning
NIC focuses on making creative innovative systems using real-world solutions and adaptations as the blueprints for those solutions
Machine learning propagates existing data/attempts using mathematical solutions to derive an answer.
