Fast and stable MAP-Elites in noisy domains using deep grids

TL;DR

This paper introduces Deep-Grid MAP-Elites, a novel variant of the MAP-Elites algorithm that uses an archive of similar solutions to improve robustness and sample efficiency in noisy domains, especially in robotics applications.

Contribution

The paper presents Deep-Grid MAP-Elites, a new approach that enhances MAP-Elites by leveraging archives of similar solutions to handle noise without extensive sampling.

Findings

More resilient to noise on behavioral descriptors

Achieves competitive fitness optimization

More sample-efficient than existing methods

Abstract

Quality-Diversity optimisation algorithms enable the evolution of collections of both high-performing and diverse solutions. These collections offer the possibility to quickly adapt and switch from one solution to another in case it is not working as expected. It therefore finds many applications in real-world domain problems such as robotic control. However, QD algorithms, like most optimisation algorithms, are very sensitive to uncertainty on the fitness function, but also on the behavioural descriptors. Yet, such uncertainties are frequent in real-world applications. Few works have explored this issue in the specific case of QD algorithms, and inspired by the literature in Evolutionary Computation, mainly focus on using sampling to approximate the "true" value of the performances of a solution. However, sampling approaches require a high number of evaluations, which in many applications such as robotics, can quickly become impractical. In this work, we propose Deep-Grid MAP-Elites, a variant of the MAP-Elites algorithm that uses an archive of similar previously encountered solutions to approximate the performance of a solution. We compare our approach to previously explored ones on three noisy tasks: a standard optimisation task, the control of a redundant arm and a simulated Hexapod robot. The experimental results show that this simple approach is significantly more resilient to noise on the behavioural descriptors, while achieving competitive performances in terms of fitness optimisation, and being more sample-efficient than other existing approaches.