R/Cextrap.R

In caRamel: Automatic Calibration by Evolutionary Multi Objective Algorithm

#' Extrapolation along orthogonal directions to the Pareto front in the space of the objectives
#'
#' gives n new candidates by extrapolation along orthogonal directions to the Pareto front in the space of the objectives
#'
#' @param param : matrix [ NPoints , NPar ] of already evaluated parameters
#' @param crit : matrix [ Npoints , NObj ] of associated criteria
#' @param directions : matrix [ NDir, 2 ] the starting and ending points of the candidate vectors
#' @param longu : matrix [ NDir , 1 ] giving the length of each segment thus defined in the OBJ space (measure of the probability of exploring this direction)
#' @param n : number of new vectors to generate
#' @return xnew : matrix [ n , NPar ] of new vectors
#' @return pcrit : matrix [ n , NObj ] estimated positions of new sets in the goal space
#'
#' @examples
#' # Definition of the parameters
#' param <- matrix(rexp(100), 100, 1)
#' crit <- matrix(rexp(200), 100, 2)
#' directions <- matrix(c(1,3,2,7,13,40), nrow = 3, ncol = 2)
#' longu <- runif(3)
#' n <- 5
#' # Call the function
#' res <- Cextrap(param, crit, directions, longu, n)
#'
#' @author Fabrice Zaoui

Cextrap <- function(param, crit, directions, longu, n) {

  ix <- rselect(n, longu)
  nobj <- dim(crit)[2]
  nvar <- dim(param)[2]

  xnew <- matrix(0, nrow = n, ncol = nvar)
  pcrit <- matrix(0, nrow = n, ncol = nobj)

  for (i in 1:n) {
    iar <- ix[i]

    param_a <- as.matrix(param[directions[iar, ], ])
    crit_a <- crit[directions[iar, ], ]

    # Search vector
    u_param <- diff(param_a)
    u_crit <- diff(crit_a)

    boost <- mean(longu) / longu[iar] # The shorter the edge, the more "boost" the search
    lambda <- -log(1 - runif(1))

    xnew[i, ] <- param_a[2, ] + boost * lambda * u_param # Coordinates in the space of the parameters
    pcrit[i, ] <- crit_a[2, ]  + boost * lambda * u_crit  # Coordinates in the space of the criteria
  }
  return(list("xnew" = xnew, "pcrit" = pcrit))
}