Want to keep learning?

This content is taken from the Taipei Medical University's online course, Applications of AI Technology. Join the course to learn more.

Skip to 0 minutes and 14 seconds Hello everybody. Today, I want to introduce Particle Swarm Optimization (PSO) and Its Applications. This is the outline of today’s course including Introduction, Particle Swarm Optimization, Applications. Now I am going to start with the Introduction. PSO was proposed by J. Kennedy an R. Eberhart in 1995. It simulates birds searching for food or the movement of fishes’ shoal. Particles in the swarm move around the search space looking for the optimum solution and adjust their position according to inertia, individual experience and social experience. Now I am going to introduce the PSO algorithm. At first we have to initialize the particles from the solution space. A particle should be with position and velocity.

Skip to 1 minute and 31 seconds In step 2, we have to Evaluate the fitness of each particle according to the fitness function. Then we can update the individual best solution PBest and the global best solution GBest. After that we update the velocity and position of each particle using these two equations. We can see here the velocity is updated according to the inertial, cognition and social experience Where omega, c1 and c2 are constants Random1 and random2 are random variables. After updating the velocity and position, go to step 2, and repeat until termination condition is reached. We can see an example for PSO solution update.

Skip to 2 minutes and 38 seconds Assume that the Current solution is (2, 2), the Particle’s best solution PBest is (2, 8), the Global best solution GBest is (7, 2),

Skip to 2 minutes and 54 seconds the Inertia: v(k) is (1, 2), omega equals to c1 equals to c2 equals to 1, random1 is 0.5, and random 2 is 0.4. Then we can obtain cognitive experience being (0,6) and social experience being (5, 0). Therefore, we have new velocity being (2, 5) and new position being (4, 7). Now we make a comparison between PSO and GA. We can see in this table that GA is Easier than PSO to find the global optimum due to “mutation”. However, the computation of GA is relatively more complicated than PSO. Finally we can see some applications of PSO. PSO can be applied for various optimization problems for example, Energy-Storage Optimization.

Skip to 4 minutes and 13 seconds Moreover, since PSO can simulate the movement of a particle swarm, this can also be applied to movie film as shown in this figure.

Particle Swarm Optimization (PSO) and its Applications

In this video, Prof. Cheng will introduce another algorithm and its applications: Particle Swarm Optimization (PSO).

Particle swarm optimization (PSO) is a robust evolutionary strategy inspired by the social behavior of animal species living in large colonies like birds, ants or fish. Prof. Cheng will present the situation of research and application in algorithm structure.

You will also see the comparison between PSO and Genetic Algorithm (GA). GA is easier than PSO to find the global optimum due to the mutation effect. However, the computation of GA is relatively more complicated than PSO.

PSO can be applied for various optimization problems, for example, Energy-Storage Optimization. PSO can simulate the movement of a particle swarm and can be applied in visual effects like those special effects in the Hollywood film. Could you tell the difference between Particle swarm optimization and Genetic Algorithm now? If not, here is an essay to tell the comparison of three evolutionary Algorithms: GA, PSO, and differential evolution(DE).

This paper talks about the general observations on the similarities and differences among the three algorithms based on computational steps. You can see the three graph Flowcharts, They will give you a simple idea of how these algorithms different from each other.

Share this video:

This video is from the free online course:

Applications of AI Technology

Taipei Medical University