Differential
Specifies the strategy used to generate trial vectors in the differential evolution algorithm.
Definition
Namespace: Numerics.NET.Optimization
Assembly: Numerics.NET (in Numerics.NET.dll) Version: 9.1.0
C#
Assembly: Numerics.NET (in Numerics.NET.dll) Version: 9.1.0
public enum DifferentialEvolutionStrategyMembers
| Best1Bin | 0 | DE/best/1/bin: Uses best solution with binomial crossover v = best + F * (r1 - r2) |
| Best1Exp | 1 | DE/best/1/exp: Uses best solution with exponential crossover v = best + F * (r1 - r2) |
| Rand1Bin | 2 | DE/rand/1/bin: Classical DE mutation with binomial crossover v = r1 + F * (r2 - r3) |
| Rand1Exp | 3 | DE/rand/1/exp: Classical DE mutation with exponential crossover v = r1 + F * (r2 - r3) |
| Rand2Bin | 4 | DE/rand/2/bin: Uses two difference vectors with binomial crossover v = r1 + F * (r2 - r3) + F * (r4 - r5) |
| Rand2Exp | 5 | DE/rand/2/exp: Uses two difference vectors with exponential crossover v = r1 + F * (r2 - r3) + F * (r4 - r5) |
| Best2Bin | 6 | DE/best/2/bin: Uses best solution and two differences with binomial crossover v = best + F * (r1 - r2) + F * (r3 - r4) |
| Best2Exp | 7 | DE/best/2/exp: Uses best solution and two differences with exponential crossover v = best + F * (r1 - r2) + F * (r3 - r4) |
| RandToBest1Bin | 8 | DE/rand-to-best/1/bin: Linear combination towards best with binomial crossover v = r1 + F * (best - r1) + F * (r2 - r3) |
| RandToBest1Exp | 9 | DE/rand-to-best/1/exp: Linear combination towards best with exponential crossover v = r1 + F * (best - r1) + F * (r2 - r3) |
| CurrentToBest1Bin | 10 | DE/current-to-best/1/bin: Linear combination from current towards best with binomial crossover v = current + F * (best - current) + F * (r1 - r2) |
| CurrentToBest1Exp | 11 | DE/current-to-best/1/exp: Linear combination from current towards best with exponential crossover v = current + F * (best - current) + F * (r1 - r2) |