Evolutionary reinforcement learning of artificial neural networks

In this article we describe EANT2, Evolutionary Acquisition of Neural Topologies, Version 2, a method that creates neural networks by evolutionary reinforcement learning. The structure of the networks is developed using mutation operators, starting from a minimal structure. Their parameters are optimised using CMA-ES, Covariance Matrix Adaptation Evolution Strategy, a derandomised variant of evolution strategies. EANT2 can create neural networks that are very specialised; they achieve a very good performance while being relatively small. This can be seen in experiments where our method competes with a different one, called NEAT, NeuroEvolution of Augmenting Topologies, to create networks that control a robot in a visual servoing scenario.