SimulatedAnnealingOptimizer.NeighborhoodSize Property

public double NeighborhoodSize { get; set; }

Public Property NeighborhoodSize As Double
	Get
	Set

public:
property double NeighborhoodSize {
	double get ();
	void set (double value);
}

member NeighborhoodSize : float with get, set

Property Value

The neighborhood size determines the extent of the search space around the current solution from which new candidate solutions are generated. A larger neighborhood size allows the algorithm to explore a wider area, potentially escaping local optima, but may result in slower convergence as the search becomes more random.

Conversely, a smaller neighborhood size focuses the search on a narrower area around the current solution, which can lead to faster convergence but increases the risk of getting trapped in local optima. The optimal neighborhood size depends on the specific problem and may require experimentation and tuning.

A good starting point for the neighborhood size is to set it to a moderate value, such as 0.1, and adjust based on the observed performance. The default value for the neighborhood size is 0.1.

Reference