|
gptkbp:instanceOf
|
mathematical optimization
|
|
gptkbp:advantage
|
simple to implement
few control parameters
may require many function evaluations
robust performance
sensitive to parameter settings
|
|
gptkbp:appliesTo
|
gptkb:machine_learning
parameter estimation
engineering optimization
|
|
gptkbp:basedOn
|
population-based search
|
|
gptkbp:category
|
evolutionary algorithm
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
differential evolution
|
|
gptkbp:introduced
|
gptkb:Rainer_Storn
gptkb:Kenneth_Price
|
|
gptkbp:introducedIn
|
1995
|
|
gptkbp:license
|
varies by implementation
|
|
gptkbp:notablePublication
|
gptkb:Differential_Evolution_–_A_Simple_and_Efficient_Heuristic_for_Global_Optimization_over_Continuous_Spaces
|
|
gptkbp:openSource
|
gptkb:DEAP
gptkb:SciPy
gptkb:PyGMO
|
|
gptkbp:parameter
|
population size
crossover rate
mutation factor
|
|
gptkbp:relatedTo
|
gptkb:particle_swarm_optimization
genetic algorithm
|
|
gptkbp:searchSpace
|
continuous
discrete (with modifications)
|
|
gptkbp:step
|
mutant
crossover
selection
initialization
|
|
gptkbp:usedFor
|
global optimization
|
|
gptkbp:uses
|
mutant
crossover
selection
|
|
gptkbp:bfsParent
|
gptkb:scipy.optimize
|
|
gptkbp:bfsLayer
|
7
|