The {ao} R package implements an iterative procedure known as alternating optimization, which optimizes a function jointly over all parameters by alternately performing restricted optimization over individual parameter subsets. For additional details on the method, please refer to the package vignette.
You can install the released version from CRAN with:
install.packages("ao")And the development version from GitHub with:
# install.packages("devtools")
devtools::install_github("loelschlaeger/ao")The following lines perform alternating optimization of the Himmelblau’s function, separately for \(x_1\) and \(x_2\), with the parameter restrictions \(-5 \leq x_1, x_2 \leq 5\):
library("ao")
himmelblau <- function(x) (x[1]^2 + x[2] - 11)^2 + (x[1] + x[2]^2 - 7)^2
ao(
f = himmelblau, p = c(0, 0), partition = list(1, 2),
base_optimizer = optimizer_optim(lower = -5, upper = 5, method = "L-BFGS-B")
)
#> $optimum
#> [1] 1.940035e-12
#>
#> $estimate
#> [1] 3.584428 -1.848126
#>
#> $sequence
#> iteration partition time p1 p2
#> 1 0 0 0.0000000000 0.000000 0.000000
#> 2 1 1 0.0211861134 3.395691 0.000000
#> 3 1 2 0.0002582073 3.395691 -1.803183
#> 4 2 1 0.0002069473 3.581412 -1.803183
#> 5 2 2 0.0001850128 3.581412 -1.847412
#> 6 3 1 0.0002598763 3.584381 -1.847412
#> 7 3 2 0.0001449585 3.584381 -1.848115
#> 8 4 1 0.0001430511 3.584427 -1.848115
#> 9 4 2 0.0001301765 3.584427 -1.848126
#> 10 5 1 0.0001349449 3.584428 -1.848126
#> 11 5 2 0.0001239777 3.584428 -1.848126
#>
#> $time
#> Time difference of 0.02668214 secsHave a question, found a bug, request a feature, want to contribute? Please file an issue.