Particle Swarm Optimizer (PSO) is a type of bio-inspired optimizationBio-inspired algorithms are a type of optimization algorithm which are modeled on biological systems. algorithm that models animal swarms like fish and flies to find the global minimum of a given optimization problem. Different agents are randomly scattered across the solution space that gradually convergeIn optimization, convergence means that the solutions returned by the algorithm start to get closer with successive iterations. Convergence is an indication that the algorithm is able to find an optimal solution for the given problem. around the optimal solution.
This algorithm is well-suited for optimizing a non-convex function that has multiple local minima. For such functions, local optimization algorithms like gradient descent optimizer will not perform very well as the optimal solution is highly dependent on the starting point.
In this project, we will implement PSO and apply it to a non-convex function.