Comparing Peircean Algorithm with Various Bio-inspired Techniques for Multi-dimensional Function Optimization

Quratulain Quraishi, Hammad Rizwan, Mian Muhammad Awais

Abstract: Peirce’s theory of evolution gives a highly efficient optimiza- tion algorithm in the domain of evolutionary computation (EC). Peircean Evolutionary Algorithm (P-EA) has the potential to solve the existing drawbacks of classical Evolutionary Algorithms such as loss of diversity, stagnation, or premature convergence. In this work, we compare P-EA with other state- of-the-art algorithms on a set of benchmark mathematical functions that are widely used to gauge their performance. These algorithms are already tested on several mathematical functions, but never have been compared with P-EA. The experimental results show that P-EA outperforms on complex functions like Michalewicz and Rastrigin with higher number of dimensions. These results help to improve the viability of P-EA for EC community.