Benchmark test functions
The following benchmark functions are implemented, each function is defined in the survey[1]:
The
Sphere
function is defined as:\[f(\mathbf{x}) = \sum_{i=1}^{d} x_i^2\]with $d$ the dimension of the design vector $\mathbf{x}$, subject to $0 \leq x_i \leq 10$.
- The minimum is\[f(\mathbf{x^*}) = 0, \quad \mathbf{x^*} = (0, \cdots, 0)\]
- The minimum is
The
Easom
function is defined as:\[f(\mathbf{x}) = -\cos{(x_1)} \cos{(x_2)} \exp{[-(x_1 - \pi)^2 - (x_2 - \pi)^2]}\]where the design vector is a 2-D vector only, subject to $-100 \leq x_i \leq 100$.
- The function has the following minimum:\[f(\mathbf{x^*}) = -1, \quad \mathbf{x^*} = (\pi, \pi)\]
- The function has the following minimum:
References
[1]
Jamil, M., & Yang, X. S. (2013). A literature survey of benchmark functions for global optimisation problems. International Journal of Mathematical Modelling and Numerical Optimisation, 4(2), 150–194. https://doi.org/10.1504/IJMMNO.2013.055204