PartitionOptimizationContextStopAtIntermediateIteration Method

protected override bool StopAtIntermediateIteration(
	int iteration,
	LinkedList<double> levels,
	LinkedList<DoubleMatrix> parameters
)

Protected Overrides Function StopAtIntermediateIteration ( 
	iteration As Integer,
	levels As LinkedList(Of Double),
	parameters As LinkedList(Of DoubleMatrix)
) As Boolean

protected:
virtual bool StopAtIntermediateIteration(
	int iteration, 
	LinkedList<double>^ levels, 
	LinkedList<DoubleMatrix^>^ parameters
) override

abstract StopAtIntermediateIteration : 
        iteration : int * 
        levels : LinkedList<float> * 
        parameters : LinkedList<DoubleMatrix> -> bool 
override StopAtIntermediateIteration : 
        iteration : int * 
        levels : LinkedList<float> * 
        parameters : LinkedList<DoubleMatrix> -> bool 

Parameters

iteration  Int32

The current iteration identifier.

levels  LinkedListDouble

The performance levels achieved in previous iterations by a SystemPerformanceOptimizer executing in this context.

parameters  LinkedListDoubleMatrix

The sampling parameters applied in previous iterations by a SystemPerformanceOptimizer executing in this context.

Return Value

At intermediate iterations, the algorithm will stop under the following conditions.

A PartitionOptimizationContext never stops before executing a number of iterations less than MinimumNumberOfIterations, and always stops if such number is greater than or equal to MaximumNumberOfIterations.

For intermediate iterations, method StopAtIntermediateIteration(Int32, LinkedListDouble, LinkedListDoubleMatrix) is called to check if a Cross-Entropy program executing in this context should stop or not.

In a PartitionOptimizationContext, the method analyzes the currently updated reference parameter, say



as follows. Define the sequence such that, for ,



If condition



can be verified, where is MinimumNumberOfIterations, the method returns true; otherwise false is returned. Equivalently, the algorithm converges if the indexes of the largest probabilities in each of the columns of the reference parameter coincide times in a row of iterations.

Reference