Group Search Optimizer (GSO) (Population base algorithm)

Group Search Optimizer (GSO) (Population base algorithm)

1. Group Search Optimizer (GSO) Dr. Ahmed Fouad Ali Faculty of Computers and Informatics Suez Canal University

2. Outline 1. Group search optimizer (GSO)(Main idea) 2. History of GSO algorithm 4. GSO Algorithm 3. Group search optimizer (GSO) 5. References

3. Group search optimizer GSO (Main idea) A group can be defined as a structured collection of interacting organisms (or members). The original idea of GSO comes from the social behavior of animals foraging and group living theory. GSO is based on Producer- Scrounger (PS) behavior of group living animals , which assume group members producing (searching for foods) and scrounging (joining resources uncovered by others).

4. History of GSO algorithm GSO algorithm is a novel swarm intelligence optimization algorithm, first published by He et al (2006) . GSO algorithm is the novel population based nature inspired algorithm, especially animal searching behavior.

5. Group search optimizer (GSO) The population of the GSO algorithm is called a group and each individual in the population is called a member . In an n-dimensional search space , the i th member at the k th searching iteration, has 1- a current position X k i ∈ R n . 2- a head angle ϕ k i = ( ϕ k i1 , . . . , ϕ k i (n−1) ) ∈ R n−1 . 3- a head direction D k i ( ϕ k i ) = ( d k i1 , . . . , d k in ) ∈ R n . which can be calculated from ϕ k i via a Polar to Cartesian coordinates Transformation:

6. Group search optimizer GSO D k i ( ϕ k i ) = ( d k i1 , . . . , d k in ) ∈ R n .

7. Group search optimizer (GSO ) (Cont.) In GSO , a group consists three kinds of members : producers and scroungers whose behaviors are based on the PS model , and rangers who perform random walk motions. The PS model is simplified by assuming that there is only one producer at each searching Iteration and the remaining members are scroungers and rangers .

8. GSO algorithm In the GSO algorithm, at the k th iteration the producer Xp behaves as follows: 1) The producer will scan at zero degree and then scan laterally by randomly sampling three points in the scanning Field as follows: Scanning field at 3D space

9. GSO algorithm (Cont.) One point at zero degree: One point in the right hand side hypercube: One point in the left hand side hypercube: where r1 ∈ R1 is a normally distributed random number with mean 0 and standard deviation 1 and r2 ∈ Rn−1 is a random sequence in the range (0, 1). Diversification (2) (3) (4)

10. The producer will then find the best point with the best resource (fitness value). If the best point has a better resource than its current position , then it will fly to this point . Or it will stay in its current position and turn its head to a new angle: Where α max is the maximum turning angle . (5) GSO algorithm (Cont.)

11. If the producer cannot find a better area after a iterations, it will turn its head back to zero degree: Where a is a constant . During each searching iteration , a number of group members are selected as scroungers . The scroungers will keep searching for opportunities to join the resources by random walk toward the producer . Where r 3 ∈ R n is a uniform random sequence in the range (0 , 1). Intensification (6) (7) GSO algorithm (Cont.)

12. Eventually, random walks, are employed by rangers . If the i th group member is selected as a ranger, at the k th iteration it generates a random head angle ϕi : where α max is the maximum turning angle; and (2) it chooses a random distance: And move to the new point (9) (8) GSO algorithm (Cont.)

13. GSO algorithm (Cont.)

14. References Computational Intelligence An Introduction Andries P. Engelbrecht , University of Pretoria South Africa S. He, Q. H. Wu, “ A Novel Group Search Optimizer Inspired by Animal Behavioural Ecology”, 2006 IEEE Congress on Evolutionary Computation Sheraton Vancouver Wall Centre Hotel, Vancouver, BC, Canada July 16-21, 2006