China
Africa
Explore chapters and articles related to this topic
Stochastic Diffusion Search: A Tutorial
Published in Adam Slowik, Swarm Intelligence Algorithms, 2020
Mohammad Majid-al-Rifaie, J. Mark Bishop
This chapter explains the principles of Stochastic Diffusion Search, a multi-agent global search and optimisation swarm intelligence algorithm based upon simple iterated interactions between agents. First a high-level description of the algorithm is presented in the form of a search metaphor driven by social interactions. This is then followed by an example of a trivial ‘text search’ application to illustrate the core algorithmic processes by which standard SDS operates.
Stochastic Diffusion Search: Modifications and Application
Published in Adam Slowik, Swarm Intelligence Algorithms, 2020
Mohammad Majid al-Rifaie, J. Mark Bishop
Stochastic Diffusion Search (SDS) [1] introduced a new probabilistic approach for solving best-fit pattern recognition and matching problems. SDS, as a multi-agent population-based global search and optimisation algorithm proposed in 1989, is a distributed mode of computation utilising interaction between simple agents [2].
Swarm Intelligence Techniques for Optimizing Problems
Published in Anand Nayyar, Dac-Nhuong Le, Nhu Gia Nguyen, Advances in Swarm Intelligence for Optimizing Problems in Computer Science, 2018
Swarm intelligence in communications can be explained on the basis of the following types: Ant colony optimization: A set of optimization algorithms that demonstrates behaviour based on the activities of an ant colony. This model is very suitable for finding the paths to goals. The simulated ants in the colony serve as simulation agents to trace optimal solutions by relocating over a parameter space searching for all the possible solutions. Real ants put down pheromones guiding each other to sources of food, while discovering their environment. The replicated ants correspondingly record their locations and the quality of their discoveries, so that in future simulation repetitions more ants find better clarifications (Ma, Sun, & Chen, 2017; Zheng, Zecchin, Newman, Maier, & Dandy, 2017).Particle swarm optimization (PSO): A global optimization methodology for handling the problems of a specific pattern, it aims to obtain the best solution among many available solutions. The advantage of this model is that it is more useful when there are a large number of members that constitute the particle swarm, making the system impressively strong to resist the problem of local minima. In the communication networks, PSO is more useful for finding the best routing that helps in the energy efficient operations (Javan, Mokari, Alavi, & Rahmati, 2017).Stochastic diffusion search (SDS): A proxy-based probabilistic universal search and optimization technique, it is suited to problems where the objective function can decompose into many independent partial functions (Kassabalidis et al., 2001). An SDS is defined based on distributed computation, where the actions of simple computational units, or agents, are inherently probabilistic. Agents supportively construct the solution by execution of autonomous searches, surveyed by the distribution of data through the population. Positive feedback encourages better results by assigning to them more agents for their investigation. Limited resources provoke strong opposition from which the major population of agents equivalent to the best-fit solution rapidly emerges.
Intelligent process of spectrum handoff for dynamic spectrum access in cognitive radio network using swarm intelligence
Published in International Journal of Computers and Applications, 2022
M. Kalpana Devi, K. Umamaheswari
A PSO is one of the optimization techniques for developing population-based algorithm. PSO was intended to create a software simulation of birds flocking the food sources, but later realized how well their algorithm gave a solution to optimization problems. PSO is an evolutionary technique like GA, stochastic diffusion search (SDS), Fuzzy logic, etc. The PU and the SU are considered as ‘particle’ in fitness value and velocity [16]. There are two ‘best’ values found during each iteration process: pbest and gbest. where is the velocity component, the is particle components, is the best particle, is the global best, and are two constant parameters, and are random numbers [0,1].
A survey on the studies employing machine learning (ML) for enhancing artificial bee colony (ABC) optimization algorithm
Published in Cogent Engineering, 2020
Dervis Karaboga, Bahriye Akay, Nurhan Karaboga
a) Biology inspired optimization (BIO) algorithms were inspired from biological phenomena or natural organisms and the algorithms existing in this group can be also classified into two subgroups which are evolutionary algorithms (EAs) inspired by the process of natural evolution and swarm intelligence (SI) based algorithms mimicking the intelligent cooperative behaviours of natural creature societies; b) the second group of NIO algorithms consists of physics-based algorithms inspired from physics systems, such as simulated annealing (SA) (Kirkpatrick et al., 1983), gravitational search (GSA) (Rashedi et al., 2009), water drop (WD) (Hosseini, 2007), harmony search (HS) (Geem et al., 2001) and stochastic diffusion search (SDS) (Bishop, 1989) algorithm.
Hybrid RNN-FFBPNN Optimized with Glowworm Swarm Algorithm for Lung Cancer Prediction
Published in IETE Journal of Research, 2023
K. Priyadarshini, Manjunathan Alagarsamy, K. Sangeetha, Dineshkumar Thangaraju
Shanthi and Rajkumar [38] predicted lung cancer utilizing a stochastic diffusion search base feature selection method with machine learning. The feature selection was suggested by using the modified stochastic diffusion search mode. The stochastic diffusion search gets gain from the direct communication to recognize optimum feature subsets. It provides maximum specificity with the maximum error rate.