| Author | Nyi, Nyi Win |
| Call Number | AIT Thesis no.CS-93-16 |
| Subject(s) | Genetic algorithms
|
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of
Engineering |
| Publisher | Asian Institute of Technology |
| Abstract | Mutation has long been regarded as a "background" operator in Genetic Algorithm, mostly
applied as a way to maintain the diversity of the population. In the adaptive mutation rate
approach, greater emphasis is put on the mutation mechanism. Mutation rates are no longer
treated as global, external parameters which are fixed or varied over time. Instead, they are
integrated into the genetic representation of individual structures and become temporal and
individually differing parameters. Moreover, they are also subject to crossover and mutation, in
a similar way as the object variables are.
Experiments are carried out to test the effectiveness of the adaptive mutation rate
mechanism. Variants of this approach are applied in the task of optimizing a set of standard test
functions and their performances are compared. It is shown that by incorporating the adaptive
mutation rates into the basic Genetic Algorithm, it is possible to achieve a more effective and
reliable search strategy. |
| Year | 1993 |
| Type | Thesis |
| School | School of Engineering and Technology (SET) |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Computer Science (CS) |
| Chairperson(s) | Yulu, Qi;Sadananda, Ramakoti; |
| Examination Committee(s) | Nagarur, Nagendra N.; |
| Scholarship Donor(s) | Deurscher Akademischer Austausch Dienst (DAAD); |
| Degree | Thesis (M.Eng.) - Asian Institute of Technology, 1993 |