Robust Optimization through Neuroevolution

TL;DR

This paper introduces a neuroevolution method that produces robust neural network solutions capable of handling environmental variations, demonstrated through multiple control and robotic tasks, outperforming existing approaches.

Contribution

It presents a novel neuroevolution approach for robust solutions, showing effectiveness and computational efficiency across diverse control problems.

Findings

Improved performance on double-pole balancing problem.

Outperformed human-designed controllers in car racing.

Generated effective solutions for swarm robotics.

Abstract

We propose a method for evolving solutions that are robust with respect to variations of the environmental conditions (i.e. that can operate effectively in new conditions immediately, without the need to adapt to variations). The obtained results show how the method proposed is effective and computational tractable. It permits to improve performance on an extended version of the double-pole balancing problem, to outperform the best available human-designed controllers on a car racing problem, and to generate rather effective solutions for a swarm robotic problem. The comparison of different algorithms indicates that the CMA-ES and xNES methods, that operate by optimizing a distribution of parameters, represent the best options for the evolution of robust neural network controllers.