TY  - CHAP
U1  - Konferenzveröffentlichung
A1  - Hagg, Alexander
A1  - Berns, Sebastian
A1  - Asteroth, Alexander
A1  - Colton, Simon
A1  - Bäck, Thomas
T1  - Expressivity of parameterized and data-driven representations in quality diversity search
T2  - Chicano, Krawiec (Eds.): GECCO '21: Genetic and Evolutionary Computation Conference, Lille, France, 10-14 July, 2021
N2  - We consider multi-solution optimization and generative models for the generation of diverse artifacts and the discovery of novel solutions. In cases where the domain's factors of variation are unknown or too complex to encode manually, generative models can provide a learned latent space to approximate these factors. When used as a search space, however, the range and diversity of possible outputs are limited to the expressivity and generative capabilities of the learned model. We compare the output diversity of a quality diversity evolutionary search performed in two different search spaces: 1) a predefined parameterized space and 2) the latent space of a variational autoencoder model. We find that the search on an explicit parametric encoding creates more diverse artifact sets than searching the latent space. A learned model is better at interpolating between known data points than at extrapolating or expanding towards unseen examples. We recommend using a generative model's latent space primarily to measure similarity between artifacts rather than for search and generation. Whenever a parametric encoding is obtainable, it should be preferred over a learned representation as it produces a higher diversity of solutions.
KW  - Generative Models
KW  - Quality diversity
KW  - Variational Autoencoder
SN  - 978-1-4503-8350-9
SB  - 978-1-4503-8350-9
U6  - https://doi.org/10.1145/3449639.3459287
DO  - https://doi.org/10.1145/3449639.3459287
AX  - 2105.04247
SP  - 678
EP  - 686
S1  - 9
PB  - Association for Computing Machinery
CY  - New York, NY, USA
ER  -