plotGameGrid {GPGame}

Visualisation of equilibrium solution in input/output space

Description

Plot equilibrium for 2 objectives test problems with evaluations on a grid. The number of variables is not limited.

Usage

plotGameGrid(
  fun = NULL,
  domain = NULL,
  n.grid,
  graphs = c("both", "design", "objective"),
  x.to.obj = NULL,
  integcontrol = NULL,
  equilibrium = c("NE", "KSE", "CKSE", "NKSE"),
  fun.grid = NULL,
  Nadir = NULL,
  Shadow = NULL,
  calibcontrol = NULL,
  ...
)

Arguments

fun

name of the function considered

domain

optional matrix for the bounds of the domain (for now [0,1]^d only), (two columns matrix with min and max)

n.grid

number of divisions of the grid in each dimension (must correspond to n.s for Nash equilibriums)

graphs

either "design", "objective" or "both" (default) for which graph to display

x.to.obj, integcontrol

see solve_game (for Nash equilibrium only)

equilibrium

either "NE" for Nash, "KSE" for Kalai-Smoridinsky and "NKSE" for Nash-Kalai-Smoridinsky

fun.grid

optional matrix containing the values of fun at integ.pts. Computed if not provided.

Nadir, Shadow

optional vectors of size nobj. Replaces the nadir point for KSE. If only a subset of values needs to be defined, the other coordinates can be set to Inf (resp. -Inf.

calibcontrol

an optional list for calibration problems, containing target a vector of target values for the objectives, log a Boolean stating if a log transformation should be used or not and offset a (small) scalar so that each objective is log(offset + (y-T^2)).

...

further arguments to fun

Value

list returned by invisible() with elements:

• trueEqdesign design corresponding to equilibrium value trueEq

• trueEqPoff corresponding values of the objective

• trueParetoFront Pareto front

• response.grid

• integ.pts, expanded.indices

Examples

library(GPareto)

## 2 variables
dom <- matrix(c(0,0,1,1),2)

plotGameGrid("P1", domain = dom, n.grid = 51, equilibrium = "NE")
plotGameGrid("P1", domain = dom, n.grid = rep(31,2), equilibrium = "NE") ## As in the tests
plotGameGrid("P1", domain = dom, n.grid = 51, equilibrium = "KSE")
plotGameGrid("P1", domain = dom, n.grid = rep(31,2), equilibrium = "NKSE")
plotGameGrid("P1", graphs = "design", domain = dom, n.grid = rep(31,2), equilibrium = "NKSE")

## 4 variables
dom <- matrix(rep(c(0,1), each = 4), 4)
plotGameGrid("ZDT3", domain = dom, n.grid = 25, equilibrium = "NE", x.to.obj = c(1,1,2,2))