Bootstrapping Conditional GANs for Video Game Level Generation
release_6uazxl2n3ncv5imokuflvqf73e
by
Ruben Rodriguez Torrado, Ahmed Khalifa, Michael Cerny Green, Niels
Justesen, Sebastian Risi, Julian Togelius
2019
Abstract
Generative Adversarial Networks (GANs) have shown im-pressive results for
image generation. However, GANs facechallenges in generating contents with
certain types of con-straints, such as game levels. Specifically, it is
difficult togenerate levels that have aesthetic appeal and are playable atthe
same time. Additionally, because training data usually islimited, it is
challenging to generate unique levels with cur-rent GANs. In this paper, we
propose a new GAN architec-ture namedConditional Embedding Self-Attention
Genera-tive Adversarial Network(CESAGAN) and a new bootstrap-ping training
procedure. The CESAGAN is a modification ofthe self-attention GAN that
incorporates an embedding fea-ture vector input to condition the training of
the discriminatorand generator. This allows the network to model
non-localdependency between game objects, and to count objects. Ad-ditionally,
to reduce the number of levels necessary to trainthe GAN, we propose a
bootstrapping mechanism in whichplayable generated levels are added to the
training set. Theresults demonstrate that the new approach does not only
gen-erate a larger number of levels that are playable but also gen-erates fewer
duplicate levels compared to a standard GAN.
In text/plain
format
Archived Files and Locations
application/pdf 489.0 kB
file_lhy2jmlevzaqtbynjxwvnchi74
|
arxiv.org (repository) web.archive.org (webarchive) |
1910.01603v1
access all versions, variants, and formats of this works (eg, pre-prints)