I ave seen the Neural network SOM equations like this:
and this PSO equation:
Where xBest and gBest denote the best particle position and best group position and the parameters ω, c1, c2, r1 and r2 are respectively inertia weight, two positive constants and two random parameters within.
In the baseline particle swarm optimisation algorithm ω is selected as unit, but an improvement of the algorithm is found in its inertial implementation using ω ≈ [0.5 0.9]. Usually maximum and minimum velocity values are also defined and initially the particles are distributed randomly to encourage the search in all possible locations.
so here said:
In other words, the weight vector is ‘moved’ closer towards the input vector.
from this part:
so the βij(t) and σ(t) coefficients are like this:
which i think maybe become like some stabilizing PSO Method ( will fount and updated ... )
So i like to know is this view point correct and can be pretended as one neural PSO dancing method to gather around one optimum answer, which is the best fitness found by winner neuron?