optimization-package {optimization}

Flexible Optimization of Complex Loss Functions with State and Parameter Space Constraints

Description

Flexible optimizer with numerous input specifications for detailed parameterisation. Designed for complex loss functions with state and parameter space constraints. Visualization tools for validation and analysis of the convergence are included.

Details

Package: optimization Type: Package Version: 1.0-6 Date: 2017-09-23 License: GPL-2

Author(s)

NA

Maintainer: NA

References

Corana, A., Marchesi, M., Martini, C. and Ridella, S. (1987), Minimizing Multimodal Functions of Continuous Variables with the 'Simulated Annealing' Algorithm. ACM Transactions on Mathematical Software, 13(3):262-280.

Gao, F. and Han, L. (2012). Implementing the nelder-mead simplex algorithm with adaptive parameters. Computational Optimization and Applications, 51(1):259 277.

Geiger, C. and Kanzow, C. (1999). Das nelder-mead-verfahren. Numerische Verfahren zur Loesung unregestrierter Optimierungsaufgaben.

Kirkpatrick, S., Gelatt, C. D. and Vecchi, M. P. (1983). Optimization by Simulated Annealing. Science, 220(4598): 671-680.

Nelder, J. and Mead, R. (1965). A simplex method for function minimization. Computer Journal, 7(4).

Pronzato, L., Walter, E., Venot, A. and Lebruchec, J.-F. (1984). A general-purpose global optimizer: Implementation and applications. Mathematics and Computers in Simulation, 26(5):412-422.

See Also

optim_nm, optim_sa, optim, plot

Examples

hi <- function(x){(x[1]**2 + x[2] - 11)**2 + (x[1] + x[2]**2 -7)**2}
optim_nm(fun = hi, k = 2)
optim_sa(fun = hi, start = c(runif(2, min = -1, max = 1)),
  trace = FALSE,
  lower = c(-4, -4),
  upper = c(4, 4),
  control = list(dyn_rf = FALSE,
    rf = 1.2,
    t0 = 10,
    nlimit = 100,
    r = 0.6,
    t_min = 0.1
  )
)