Adaptation in a System Metamodel for Evolutionary Computation

Complex systems are difficult to model and analyse with traditional mathematics, e.g. analytically solving differential equations. To better understand such systems, we recently presented a system metamodel to provide an algorithmic alternative to analytical methods and were able to create simulations of complex systems modelled as cellular automata and artificial neural networks. In this study we extend our system metamodel with the concept of adaption in order to integrate evolutionary computation in our so-called allagmatic method. Adaption is described in the context of our system metamodel and the allagmatic method, defined and implemented, and computational experiments with cellular automata and artificial neural networks performed. We find that the system metamodel of the allagmatic method integrates adaptation with an additional operation called adaptation function that operates on the update function, which encodes the system's dynamics. It allows the creation of evolutionary computations by providing an abstract template for adaptation and guidance for implementation with the system metamodel. The creation of the system metamodel was first inspired by abstract concepts of the philosophy of individuation of Gilbert Simondon. The theoretical background for the concept of adaptation in this study is taken from the philosophy of organism of Alfred North Whitehead.