Semantic Mutation Operator for Fast and Efficient Design of Bent Boolean Functions

Bent functions are a type of Boolean functions with properties that make them useful in cryptography. In this paper we propose a new semantic mutation operator for design of bent Boolean functions via genetic programming. To assess the efficiency of the proposed operator, we compare it to several other commonly used non-semantic mutation operators. Our results show that semantic mutation makes the evolutionary process more efficient, and significantly decreases the number of fitness function evaluations required to find a bent function.