Multimodal Optimization by Sparkling Squid Populations
It introduces an incremental method for optimization problems in fields like computer vision and logistics.
The paper proposes the Sparkling Squid Algorithm (SSA), a new swarm intelligence method for multimodal optimization inspired by squid behavior, and compares it to existing metaheuristics with applications to problems like image registration and the traveling salesperson problem.
The swarm intelligence of animals is a natural paradigm to apply to optimization problems. Ant colony, bee colony, firefly and bat algorithms are amongst those that have been demonstrated to efficiently to optimize complex constraints. This paper proposes the new Sparkling Squid Algorithm (SSA) for multimodal optimization, inspired by the intelligent swarm behavior of its namesake. After an introduction, formulation and discussion of its implementation, it will be compared to other popular metaheuristics. Finally, applications to well - known problems such as image registration and the traveling salesperson problem will be discussed.