DErand

DE/rand/k variant for discriminative Bayesian Network parameter learning

DErand implements the DE/rand/1 and DE/rand/2 mutation strategies. Unlike DEbest, the base vector is randomly selected from the population rather than being the current best individual — promoting broader exploration.

Usage

DErand(
  NP = 40,
  G = 100,
  data,
  class.name,
  F = 0.5,
  CR = 0.7,
  mutation.pairs = c(1, 2),
  crossover = c("bin", "exp"),
  structure = c("nb", "tancl", "tan", "hc"),
  edgelist = NULL,
  verbose = 25,
  ...
)

Parameters

NP · integer · default: "40"

Population size. Must be > 5; values ≤ 5 are reset to 40.

G · integer · default: "100"

Maximum generations. Values ≤ 1 are reset to 100.

data · data.frame · required

Data frame from which to learn the classifier.

class.name · character · required

Name of the class variable column in data.

F · numeric · default: "0.5"

Mutation factor in [0, 2].

CR · numeric · default: "0.7"

Crossover rate in [0, 1].

mutation.pairs · integer · default: "1"

Number of difference vector pairs: 1 (DE/rand/1) or 2 (DE/rand/2).

crossover · character · default: "bin"

Crossover type: "bin" (binomial) or "exp" (exponential).

structure · character · default: "nb"

BN structure: "nb", "tan", "tancl", or "hc".

edgelist · matrix

Optional custom topology as an edges × 2 matrix (from, to).

verbose · integer · default: "25"

Print progress every this many generations. 0 suppresses output.

...

Extra arguments forwarded to bnclassify::tan_cl or bnclassify::tan_hc.

Return value

Returns an object of class DE with the same fields as DEbest: Best, BestCLL, pobFinal, CLLPobFinal, N.evals, convergence, evaluations.

DErand vs DEbest

Aspect	DEbest	DErand
Base vector	Always the best individual	Random population member
Exploration	Less — converges faster	More — less prone to premature convergence
Exploitation	Stronger	Weaker
Best for	Unimodal or well-behaved landscapes	Multimodal landscapes

Examples

library(dplbnDE)
data(car)

# DE/rand/2/bin with TAN structure
result <- DErand(
  NP = 20, G = 25,
  data = car,
  class.name = names(car)[7],
  F = 0.5, CR = 0.5,
  mutation.pairs = 2,
  crossover = "bin",
  structure = "tan",
  verbose = 5
)

print(result)
predictions <- predict(result$Best, car)
accuracy(predictions, car$class)