7. Parameters and Parameter Tuning

Chapter 3 presented an algorithmic framework that forms the common basis for all evolutionary algorithms. A decision to use an evolutionary algorithm implies that the user adopts the main design decisions behind this framework. Thus, the main algorithm setup follows automatically: the algorithm is based on a population of candidate solutions that is manipulated by selection, recombination, and mutation operators. To obtain a concrete, executable EA, the user only needs to specify a few details.

In this chapter we have a closer look at these details, named parameters. We discuss the notion of EA parameters and explain why the task of designing an evolutionary algorithm can be seen as the problem of finding appropriate parameter values. Furthermore, we elaborate on the problem of tuning EA parameters and provide an overview of different algorithms that can tune EAs with limited user effort.

Content:

  1. 7.1  Evolutionary Algorithm Parameters……………………119
  2. 7.2  EAs and EA Instances ……………………………..120
  3. 7.3  Designing Evolutionary Algorithms ……………………121
  4. 7.4  The Tuning Problem……………………………….123
  5. 7.5  Algorithm Quality: Performance and Robustness . . . . . . . . . . . . 125
  6. 7.6  Tuning Methods…………………………………..128

Suggested Reading

Exercises

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

The on-line accompaniment to the book Introduction to Evolutionary Computing