Kaipa, Krishnanand N and Ghose, Debasish (2017) Theoretical Foundations. [Book Chapter]
Full text not available from this repository.Abstract
The previous chapter described the development of GSO and its basic ability to optimize multimodal functions. This chapter deals with the theoretical foundations for GSO. Initially, a characterization of the swarming behavior of agents in GSO is carried out, which guides the formulation of a framework used to analyze the algorithm. A theoretical model of GSO is obtained by using assumptions that make it amenable to analysis, yet reflecting most of the features of the original algorithm. Next, local convergence results for this GSO model are provided: First, when the glowworms are under the influence of an isolated leader, an upper bound on the time taken by the glowworms to converge to the isolated leader is determined. Second, when the glowworms are under the influence of two leaders with non-isolated and non-overlapping neighborhoods, an upper bound on the time taken by the glowworms to converge to one of the leaders is determined. Third, it is shown that glowworms, under the influence of two leaders with overlapping neighborhoods, asymptotically converge to one of the leaders. Some illustrative simulations are presented to support these theoretical findings.
| Item Type: | Book Chapter |
|---|---|
| Series.: | Studies in Computational Intelligence |
| Publisher: | Springer Verlag |
| Additional Information: | The Copyright of this article belongs to the Springer |
| Keywords: | Peak Location; Local Convergence; Function Profile; Initial Placement; Multimodal Function |
| Department/Centre: | Division of Mechanical Sciences > Aerospace Engineering(Formerly Aeronautical Engineering) |
| Date Deposited: | 30 May 2022 05:09 |
| Last Modified: | 30 May 2022 11:17 |
| URI: | https://eprints.iisc.ac.in/id/eprint/72584 |
Actions (login required)
 |
View Item |
