makeBiObjBBOBFunction: Generate Bi-Objective Function from the Real-Parameter...

In jakobbossek/smoof: Single and Multi-Objective Optimization Test Functions

Generate Bi-Objective Function from the Real-Parameter Bi-Objective Black-Box Optimization Benchmarking (BBOB)

Description

Generator for the function set of the real-parameter Bi-Objective Black-Box Optimization Benchmarking (BBOB) with Function IDs 1-55, as well as its extended version (bbob-biobj-ext) with Function IDs 1-92.

Usage

makeBiObjBBOBFunction(dimensions, fid, iid)

Arguments

dimensions	[integer(1)] Problem dimensions. Integer value between 2 and 40.
fid	[integer(1)] Function identifier. Integer value between 1 and 92.
iid	[integer(1)] Instance identifier. Integer value greater than or equal 1.

Value

[smoof_multi_objective_function] Bi-objective function from the BBOB benchmark.

Note

Concatenation of single-objective BBOB functions into a bi-objective problem.

References

See the COCO website for a detailed description of the bi-objective BBOB functions. An overview of which pair of single-objective BBOB functions creates which of the 55 bi-objective BBOB functions can be found in the official documentation of the bi-objective BBOB test suite. And a full description of the extended suite with 92 functions can be found in the official documentation of the extended bi-objective BBOB test suite.

Examples

# get the fifth instance of the concatenation of the
# 3D versions of sphere and Rosenbrock
fn = makeBiObjBBOBFunction(dimensions = 3L, fid = 4L, iid = 5L)
fn(c(3, -1, 0))
# compare to the output of its single-objective pendants
f1 = makeBBOBFunction(dimensions = 3L, fid = 1L, iid = 2L * 5L + 1L)
f2 = makeBBOBFunction(dimensions = 3L, fid = 8L, iid = 2L * 5L + 2L)
identical(fn(c(3, -1, 0)), c(f1(c(3, -1, 0)), f2(c(3, -1, 0))))

jakobbossek/smoof documentation built on Feb. 17, 2024, 2:23 a.m.