Follow us on Twitter
Lab of Misfits

This is part of the Robotsprogramme

Back

 

Analysing the evolvability of neural network agents through structural mutations


This paper investigates evolvability of artificial neural networks within an artificial life environment. Five different structural mutations are investigated, including adaptive evolution, structure duplication, and incremental changes. The total evolvability indicator, Etotal, and the evolvability function through time, are calculated in each instance, in addition to other functional attributes of the system. The results indicate that incremental modifications to networks, and incorporating an adaptive element into the evolution process itself, significantly increases neural network evolvability within open-ended artificial life simulations.

Evolvability in artificial neural networks image

Download paper


Evolutionary Benefits of Evolvable Component Integration.
Download pdf