Particle Swarm Optimization Algorithm with Reverse-Learning and Local-Learning Behavior

Abstract

In order to resolve conflict between convergence speed and population diversity of particle swarm optimization (PSO) algorithm, an improved PSO, called reverse-learning and local-learning PSO (RLPSO) algorithm, is presented in which a reverse-learning behavior implemented by some particles while local-learning behavior adopted by elite particles in each generation. During the reverse-learning process, some inferior particles of initial population and each particle's historical worst position are reserved to attract a particle to leap out of local optimums. Furthermore, the Hamming distance between the inferior particles is set to no less than a default rejection distance the aim of which is to maintain the diversity of population and improve RLPSO's exploration ability. In each generation, the difference between the best particle and the second-best particle is used to guide the best one to carry out local search which is crucial for improving RLPSO's exploitation ability. The results of experiments show that RLPSO has a good global searching ability and convergence speed especially in high dimension function.

Beetle swarm optimisation for solving investment portfolio problems

“…At the same time, one‐year national debt is used as risk‐free assets (a yield of 1.95%) on the assumption that a total of 1 million assets can be used for investment and no securities have been purchased before investment [21, 22].…”

Chicken Swarm Optimization Based on Elite Opposition‐Based Learning

“…With increase of , random search of distribution transits gradually from that of Cauchy distribution to that of Gaussian distribution, and random search of distribution is similar to that of Gaussian distribution with good ability of local development in the later stage. Tizhoosh [19] is a new technology applied to intelligent computing area, and it has been successfully applied in many intelligent algorithm optimizations [20][21][22]. Theoretically verified by Zhong and others [23], opposition-based learning can acquire a solution that closes to global optimal solution with a higher probability.…”

A Novel Power Amplifier Behavior Modeling Based on RBF Neural Network with Chaos Particle Swarm Optimization Algorithm

“…The performance of RBF network is determined by the parameters of the network, which is the center and variance of the basis function as well as network weights [10]. If the chaotic particle swarm algorithm is used for neural network training, then it is not easy for particles to get into the local optimum [11][12]. This algorithm can also expand the search space, search the global optimal solution, and speed up the fitting rate of the neural network training algorithm.…”