8. Optimization by Evolutionary Computation

In this chapter the state of the art in genetic algorithms is presented. Two of the optimizers available in the SIESTA framework (cf. Section 9.3) are based on the theory discussed here. The use of evolutionary computation or genetic algorithm optimizers for optimization and inverse modeling problems in TCAD is novel and advantageous because they complement previously used local optimizers. Finally the advantages and disadvantages of local optimizers and evolutionary computation ones are discussed.

- 8.1 Introduction
- 8.2 Representation of the Variables
- 8.3 Operators for Continuous Variables

- 8.4 How Genetic Algorithms Work
- 8.4.1 Evolutionary Computation
- 8.4.2 The Classic Genetic Algorithm
- 8.4.3 A State of the Art Algorithm

- 8.5 Schemata and the Schema Theorem
- 8.6 Comparison of Local Optimizers and Evolutionary Computation
- 8.6.1 Advantages and Disadvantages of Local Optimizers
- 8.6.2 Advantages and Disadvantages of Evolutionary Algorithm

Optimizers

- 8.7 Examples