Particle swarm optimisation with gradually increasing directed neighbourhoods

Abstract

Particle swarm optimisation (PSO) is an intelligent random search algorithm, and the key to success is to effectively balance between the exploration of the solution space in the early stages and the exploitation of the solution space in the late stages. This paper presents a new dynamic topology called "gradually increasing directed neighbourhoods (GIDN)" that provides an effective way to balance between exploration and exploitation in the entire iteration process. In our model, each particle begins with a small number of connections and there are many small isolated swarms that improve the exploration ability. At each iteration, we gradually add a number of new connections between particles which improves the ability of exploitation gradually. Furthermore, these connections among particles are created randomly and have directions. We formalise this topology using random graph representations. Experiments are conducted on 31 benchmark test functions to validate our proposed topology. The results show that the PSO with GIDN performs much better than a number of the state of the art algorithms on almost all of the 31 functions.

CFSO3: A New Supervised Swarm-Based Optimization Algorithm

“…A large series of changes from the original PSO have been proposed in order to improve its performances. In particular, many works focused on the way to manage the tuning of parameters for achieving better convergence to the global optimum and/or for improving exploration for multimodal problems (e.g., see [10][11][12][13][14][15][16][17][18][19] and the reference within), and the effects of topological rules on performances have been discussed [20][21][22][23][24][25][26]. The basic idea of topological rules is to link each member of the swarm with others by generating more complex information exchange among particles than the simple use of the global best.…”

Fully Informed Swarm Optimization Algorithms: Basic Concepts, Variants and Experimental Evaluation

“…WARSAW, 2014 systems optimization [15]. It was also used as a starting point for other similar approaches [16], [17] as well as a component of hybrid algorithms [18]. Here we consider most general FIPSO with fully-connected particles to study its performance when referencing it to two other more recent approaches.…”

Particle Swarm Optimisation for Edge Detection in Noisy Images

“…• investigate the performance of CanPSO, BBPSO and FIPS on the detection of edges in noisy images when they are equipped with different dynamic topologies (gradually increasing directed neighbourhood (GIDN) [190] and random dynamic topology [191]), and…”

“…A topology can control the speed of information flow among particles, which can affect the exploration and exploitation abilities of PSO. Therefore, a dynamic topology, which changes the connection structure between particles over the PSO iterations, can control the exploration and exploitation abilities of PSO [190]. In the fully connected topology, all particles are connected to each other and accordingly they share their acquired information among all others very quickly in one iteration.…”

“…The ring topology could improve the exploration ability of the PSO algorithm but it might reduce the the convergence speed of the algorithm [115]. Therefore, a few dynamic topologies have been proposed to effectively control the exploration and exploitation abilities of PSO over its iterations [190].…”