Skip to Main Content
Table 2:
Test problems for benchmarking the niching methods, directly adapted from the CEC2013 special session on multimodal optimization (Li et al., 2013b). The niche radius parameter is used only for post-processing of the results.
PIDFunctionDimensionNiche radiusMaxEvals
Five-Uneven-Peak Trap 0.01 50,000 
Equal Maxima 0.01 50,000 
Uneven Decreasing Maxima 0.01 50,000 
Himmelblau 0.01 50,000 
Six-Hump Camel Back 0.5 50,000 
Shubert 0.5 18 200,000 
Vincent 0.2 36 200,000 
Shubert 0.5 81 400,000 
Vincent 0.2 216 400,000 
10 Modified Rastrigin 0.01 12 200,000 
11 Composition Function 1 0.01 200,000 
12 Composition Function 2 0.01 200,000 
13 Composition Function 3 0.01 200,000 
14 Composition Function 3 0.01 400,000 
15 Composition Function 4 0.01 400,000 
16 Composition Function 3 0.01 400,000 
17 Composition Function 4 0.01 400,000 
18 Composition Function 3 10 0.01 400,000 
19 Composition Function 4 10 0.01 400,000 
20 Composition Function 4 20 0.01 400,000 
PIDFunctionDimensionNiche radiusMaxEvals
Five-Uneven-Peak Trap 0.01 50,000 
Equal Maxima 0.01 50,000 
Uneven Decreasing Maxima 0.01 50,000 
Himmelblau 0.01 50,000 
Six-Hump Camel Back 0.5 50,000 
Shubert 0.5 18 200,000 
Vincent 0.2 36 200,000 
Shubert 0.5 81 400,000 
Vincent 0.2 216 400,000 
10 Modified Rastrigin 0.01 12 200,000 
11 Composition Function 1 0.01 200,000 
12 Composition Function 2 0.01 200,000 
13 Composition Function 3 0.01 200,000 
14 Composition Function 3 0.01 400,000 
15 Composition Function 4 0.01 400,000 
16 Composition Function 3 0.01 400,000 
17 Composition Function 4 0.01 400,000 
18 Composition Function 3 10 0.01 400,000 
19 Composition Function 4 10 0.01 400,000 
20 Composition Function 4 20 0.01 400,000 
Close Modal

or Create an Account

Close Modal
Close Modal