Empirical attainment function (EAF) difference

Plot the differences between the empirical attainment functions (EAFs) of two approximations sets side by side adopting ggplot2. The left plot shows the the values of the left EAF minus the right EAF and the right side shows it the other way around.

plot_eaf_diff(df1, df2, obj.cols, intervals = 5L, percentiles = c(0, 50, 75))

Arguments

df1	[data.frame] First approximation set: a data frame with columns at least those given via parameter obj.cols, and “repl”. All other columns will be ignored!
df2	[data.frame] First approximation set: a data frame with columns at least those given via obj.cols, and “repl”. All other columns will be ignored!
obj.cols	[character(>= 2)] Column names of the objective function values. Default is c("y1", "y2").
intervals	[integer(1)] The absolute range of the differences \([0, 1]\) is partitioned into the number of intervals provided. Default is 5.
percentiles	[numeric] Percentiles of the EAF that will be plotted as attainment surfaces.

Value

A ggplot object.

Note

If the given approximation sets are huge in size, generating the plot may take some time. Here, it is most often helpful to store the plot first in a file. Due to the huge number of points, the size of the file may become also very large. Here, it is advisible to save the file in an image file format (PNG or JPEG) instead of PDF.

This function re-implements partial functionality of eafdiffplot realizing the plotting with ggplot2 rather than base R plot functions.

References

[1] V. Grunert da Fonseca and C. M. Fonseca, The attainment-function approach to stochastic multiobjective optimizer assessment and comparison, in Experimental Methods for the Analysis of Optimization Algorithms (T. Bartz-Beielstein, M. Chiarandini, L. Paquete, and M. Preuss, eds.), ch. 5, pp. 103-130, Springer Berlin Heidelberg, 2010.

[2] Manuel López-Ibáñez, Luís Paquete, and Thomas Stützle. Exploratory Analysis of Stochastic Local Search Algorithms in Biobjective Optimization. In T. Bartz-Beielstein, M. Chiarandini, L. Paquete, and M. Preuss, editors, Experimental Methods for the Analysis of Optimization Algorithms, pages 209–222. Springer, Berlin, Germany, 2010. doi: 10.1007/978-3-642-02538-9_9.

[3] Knowles, J. D., Thiele, L. and Zitzler, E. A tutorial on the performance assessment of stochastive multiobjective optimizers. TIK-Report No. 214, Computer Engineering and Networks Laboratory, ETH Zurich, February 2006 (Revised version. First version, January 2005). doi: 10.3929/ethz-b-000023822.

See also

plot_eaf

Other multi-objective visualizations: plot_eaf(), plot_heatmap(), plot_pcp(), plot_radar(), plot_scatter2d(), plot_scatter3d()

Examples

if (FALSE) {
data(emoas_on_zdt)
df1 = emoas_on_zdt[emoas_on_zdt$algorithm == "nsga2"
  & emoas_on_zdt$problem == "zdt1_2d_2o", ]
df2 = emoas_on_zdt[emoas_on_zdt$algorithm == "smsemoa"
  & emoas_on_zdt$problem == "zdt1_2d_2o", ]
plot_eaf_diff(df1, df2, obj.cols = c("y1", "y2"))
}