Optimization

• Animation of particle swarm optimization algorithm used to train the neural network. Particles get initialized by being assigned to any random position and velocity and potential solution which are flown through the hyperspace.
Aplied in financial portfolio management, optimization problems, and as trading strategy algorithm.

• All particles are working towards the Global Best Solution and keeping record of each position they have taken and achieved before (Personal Best).

• Essentially iteration of each particle causes acceleration towards their own Personal Best position and overall Global Best position. These two values are weighted randomly and product is new velocity for the particles which will affect the future position.

Plot of swarm reaching Global Best Solution for 12 different parameters.

Category:

• Optimization, Pricing