Quality-Diversity Meta-Evolution: customising behaviour spaces to a meta-objective
Quality-Diversity Meta-Evolution: customising behaviour spaces to a meta-objective
Quality-Diversity (QD) algorithms evolve behaviourally diverse and high-performing solutions. To illuminate the elite solutions for a space of behaviours, QD algorithms require the definition of a suitable behaviour space. If the behaviour space is high-dimensional, a suitable dimensionality reduction technique is required to maintain a limited number of behavioural niches. While current methodologies for automated behaviour spaces focus on changing the geometry of the behaviour space or on unsupervised learning of its key features, there remains a need for customising behavioural diversity to a particular meta-objective specified by the end-user. In the newly emerging framework of QD Meta-Evolution, or QD-Meta for short, one evolves a population of QD algorithms, each with different algorithmic and representational characteristics, to optimise the algorithms and their resulting archives to a user-defined meta-objective. Despite promising results compared to traditional QD algorithms, QD-Meta has yet to be compared to state-of-the-art behaviour space automation methods such as Centroidal Voronoi Tessellations Multi-dimensional Archive of Phenotypic Elites (CVT-MAP-Elites) and Autonomous Robots Realising their Abilities (AURORA). This paper performs an empirical study of QD-Meta on function optimisation and multilegged robot locomotion benchmarks. Results demonstrate that QD-Meta archives provide improved average performance and faster adaptation to a priori unknown changes to the environment when compared to CVT-MAP-Elites and AURORA. A qualitative analysis shows how the resulting archives are tailored to the meta-objectives provided by the end-user.
Bossens, David
633a4d28-2e59-4343-98fe-283082ba1873
Tarapore, Danesh
fe8ec8ae-1fad-4726-abef-84b538542ee4
Bossens, David
633a4d28-2e59-4343-98fe-283082ba1873
Tarapore, Danesh
fe8ec8ae-1fad-4726-abef-84b538542ee4
Bossens, David and Tarapore, Danesh
(2022)
Quality-Diversity Meta-Evolution: customising behaviour spaces to a meta-objective.
IEEE Transactions on Evolutionary Computation.
(doi:10.1109/TEVC.2022.3152384).
(In Press)
Abstract
Quality-Diversity (QD) algorithms evolve behaviourally diverse and high-performing solutions. To illuminate the elite solutions for a space of behaviours, QD algorithms require the definition of a suitable behaviour space. If the behaviour space is high-dimensional, a suitable dimensionality reduction technique is required to maintain a limited number of behavioural niches. While current methodologies for automated behaviour spaces focus on changing the geometry of the behaviour space or on unsupervised learning of its key features, there remains a need for customising behavioural diversity to a particular meta-objective specified by the end-user. In the newly emerging framework of QD Meta-Evolution, or QD-Meta for short, one evolves a population of QD algorithms, each with different algorithmic and representational characteristics, to optimise the algorithms and their resulting archives to a user-defined meta-objective. Despite promising results compared to traditional QD algorithms, QD-Meta has yet to be compared to state-of-the-art behaviour space automation methods such as Centroidal Voronoi Tessellations Multi-dimensional Archive of Phenotypic Elites (CVT-MAP-Elites) and Autonomous Robots Realising their Abilities (AURORA). This paper performs an empirical study of QD-Meta on function optimisation and multilegged robot locomotion benchmarks. Results demonstrate that QD-Meta archives provide improved average performance and faster adaptation to a priori unknown changes to the environment when compared to CVT-MAP-Elites and AURORA. A qualitative analysis shows how the resulting archives are tailored to the meta-objectives provided by the end-user.
Text
paper
- Accepted Manuscript
Restricted to Repository staff only until 17 February 2024.
Request a copy
More information
Accepted/In Press date: 17 February 2022
Identifiers
Local EPrints ID: 456255
URI: http://eprints.soton.ac.uk/id/eprint/456255
ISSN: 1089-778X
PURE UUID: edd57914-90a1-4588-ac40-4181a24fd240
Catalogue record
Date deposited: 26 Apr 2022 23:52
Last modified: 25 Jun 2022 01:59
Export record
Altmetrics
Contributors
Author:
David Bossens
Author:
Danesh Tarapore
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics