Evolutionary Computing

Intelligence is the capability of a system to adapt its behaviour to an ever-changing environment.

According to Alan Turing (Turing, 1950), the form or appearance of a system is irrelevant to its intelligence. However, from our everyday experience, we know that evidences of intelligent behaviour are easily observed in humans. But we are products of evolution, and thus by modelling the process of evolution, we might expect to create intelligent behaviour. Evolutionary computation simulates evolution on a computer.

The result of such a simulation is a series of optimization algorithms, usually based on a simple set of rules. Optimization iteratively improves the quality of solutions until an optimal, or at least feasible, solution is found.

Evolutionary computing algorithms are most commonly used in optimization problems, with a given objective function and a set of constraints. For example:

The goal is to mimic evolution to find the model, algorithm, design etc. that suits the domain the best (fittest), so compiles with the restrictions and fits the requirements. Every candidate in the solution space will have this fitness, and similar solutions will have similar fitness and fitness may be subject to change.

EC algorithms are flexible optimization algorithms.

• No constraints in the type of variables.
• Works well with non-linear objective functions.
• Insensitive to discontinuities or shape.
• Can be adapted to multi-objective optimization problems.
• In general, requires few assumptions about the problem.

But is evolution really intelligent? We can consider the behaviour of an individual organism as an inductive inference about some yet unknown aspects of its environment (Fogel et al., 1966). Then if, over successive generations, the organism survives, we can say that this organism is capable of learning to predict changes in its environment. Evolution is a tortuously slow process from the human perspective, but the simulation of evolution on a computer does not take billions of years!

The evolutionary approach to machine learning is based on computational models of natural selection and genetics. We call them evolutionary computation, an umbrella term that combines genetic algorithms, evolution strategies and genetic programming. All these techniques simulate evolution by using the processes of selection, mutation, and reproduction.

Application areas of EC:

• Economics.
  • For example, to characterize economic models, asset pricing, etc.
• Logistics.
  • For example, in scheduling and travelling salesman problem (TSP).