artificial fish swarm optimization

21
Company LOGO Scientific Research Group in Egypt (SRGE) Artificial fish swarm algorithm Dr. Ahmed Fouad Ali Suez Canal University, Dept. of Computer Science, Faculty of Computers and informatics Member of the Scientific Research Group in Egypt Workshop on Swarms optimization, 6 June 2015 Ain shams university

Upload: ahmed-fouad

Post on 28-Jul-2015

108 views

Category:

Education


4 download

TRANSCRIPT

Company

LOGO

Scientific Research Group in Egypt (SRGE)

Artificial fish swarm algorithm

Dr. Ahmed Fouad AliSuez Canal University,

Dept. of Computer Science, Faculty of Computers and informatics

Member of the Scientific Research Group in Egypt

Workshop on Swarms optimization, 6 June 2015 Ain shams university

Company

LOGO Scientific Research Group in Egyptwww.egyptscience.net

Company

LOGO Outline

1. Basic concepts 1. Basic concepts

2. The fish swarm behavior 2. The fish swarm behavior

3. Artificial fish swarm Algorithm (AFSA)3. Artificial fish swarm Algorithm (AFSA)

4. AFSA: Pros and cons4. AFSA: Pros and cons

5. References 5. References

Company

LOGO Basic concepts (Meta-heuristics)

Company

LOGO Basic concepts (Meta-heuristics)

The term “meta-heuristics" was first proposed by Glover (1986).

• Meta-heuristics are global search methods that cover all heuristics methods that show evidence of achieving good quality solutions for the problem of interest within an acceptable time.

• Meta-heuristics structures are mainly based on simulating natureand artificial intelligence tools.

Company

LOGOBasic concepts (exploration and

exploitation The main two components of a meta-heuristic method are:♦ Exploration (Diversification) Process. Exploring the search space and avoiding trapping in local minima.

♦ Exploitation (Intensification) Process. Improving any promising solution obtained so far.

• Meta-heuristics can be classified into two classes:♦ Population-based methods.♦ Point-to-point methods.

Company

LOGO Basic concepts (Swarm intelligence )

• Suppose you and a group of friends are on a treasure finding mission. Each one in the group has a metal detector and can communicate the signal and current position to the n nearest neighbors.

• Each person therefore knows whether one of his neighbors is nearer to the treasure than him. If this is the case, you can move closer to that neighbor. In doing so, your chances are improved to find the treasure. Also, the treasure may be found more quickly than if you were on your own.

Company

LOGO Swarm intelligence (Main Idea)

• A swarm can be defined as a structured collection of interacting organisms (or agents).

• Within the computational study of swarm intelligence, individual organisms have included ants, bees, wasps, termites, fish (in schools) and birds (in flocks).

Company

LOGO The fish swarm behaviorRandom behavior

Searching behavior

Swarming behavior

Chasing behavior

Leaping behavior

Company

LOGOArtificial fish swarm optimization Algorithm (AFSA)• Artificial fish swarm AFSO was first

proposed in 2002 (Li et al.).

• The AFSO is a population based algorithm.

• The main issue of the artificial fish swarm algorithm is the visual scope of each fish.

• Let npi visual be the number of points in its visual scope.

Company

LOGOArtificial fish swarm optimization Algorithm (AFSA)• There are three possible situations

may occur:

• When npi visual = 0, the visual scope is empty, and the point xi, with no other points in its neighborhood to follow, moves randomly searching for a better region.

• When the visual scope is crowded, the point has some difficulty in following any

particular point, and searches for a better region choosing randomly another point(from the visual scope) and moves towards it.

Company

LOGOArtificial fish swarm optimization Algorithm (AFSA)• When the visual scope is not

crowded, the point is able either to swarm moving

towards the central or to chase moving towards the best point.

• The condition that decides when the visual scope of xi is not crowded is

Where m is the population size numberθ is crowded parameter

Company

LOGOArtificial fish swarm optimization Algorithm (AFSA)• The swarming behavior is

characterized by a movement towards the central of the

points in the visual scope of xi.

• The central point is then defined by

• The swarming movement is activated only if the central point has a better function value when compared with f(xi).

• Otherwise, the point xi randomly chooses a point inside the visual scope and moves towards it if it has a better function value. This is the searching behavior.

Company

LOGOArtificial fish swarm optimization Algorithm (AFSA)

• The chasing behavior is carried out when the minimum function value inside the visual scope of xi satisfies

Where "min" denotes the index of the point with the least function value.

• If the condition is not satisfied then the algorithm activates the searching behavior

Company

LOGO (AFSA)Algorithm Parameter setting

Initial population

Random behavior

Swarm behavior

Chase behavior

Greedy selection

Leap behavior

Company

LOGO AFSA (Random)

Company

LOGO AFSA (Moving)

Company

LOGO AFSA (Leaping )When the best objective function value in the population does not change for a certain number of iterations, the algorithm may fall into a local minimum. ("stagnation“)

Company

LOGO AFSA: Pros and cons

Pros:Global search ability

Tolerance of parameter setting

Good Robustness

Cons:Higher time complexity

Lower convergence speed

Lack of balance between global search and local search

Not use of the experiences of group members for the next moves.

Company

LOGO References

• Andries P. Engelbrecht, Computational Intelligence An Introduction,, University of Pretoria South Africa

• E. M. G. P. Fernandes, T. F. M. C. Martins and A. Rocha, Fish Swarm Intelligent Algorithm for Bound Constrained Global Optimization, Proceedings of the International Conference on Computational and Mathematical Methods in Science and Engineering, CMMSE 2009.

Company

LOGO

Thank youThank you

http://www.egyptscience.net

[email protected]