1. Convergence to a local optimum has been analyzed for PSO in and. 2. It is not guaranteed to even be a local optimum . 3. It might instead find a local optimum value, not the global optimum. 4. The algorithm ends after reaching some local optimum . 5. This will converge to a local optimum , so multiple runs may produce different results. 6. It has been proven that PSO need some modification to guarantee to find a local optimum . 7. An optimal solution is one that is a local optimum , but possibly not a global optimum. 8. Finding an arbitrary local optimum is relatively straightforward by using classical " local optimization " methods. 9. Gradient-based methods find local optima with high reliability but are normally unable to escape a local optimum . 10. It does however only find a local optimum , and is commonly run multiple times with different random initializations.
Mobile