Controllable Game Level Generation: Assessing the Effect of Negative Examples in GAN Models

TL;DR

This paper evaluates how negative examples influence controllable GANs, specifically CGAN and Rumi-GAN, in generating game levels that meet specific constraints like playability and controllability.

Contribution

It provides an empirical comparison of CGAN and Rumi-GAN in using negative examples to improve controllability in game level generation.

Findings

Negative examples improve controllability in some cases

Rumi-GAN effectively leverages negative examples

CGAN shows mixed results with negative examples

Abstract

Generative Adversarial Networks (GANs) are unsupervised models designed to learn and replicate a target distribution. The vanilla versions of these models can be extended to more controllable models. Conditional Generative Adversarial Networks (CGANs) extend vanilla GANs by conditioning both the generator and discriminator on some additional information (labels). Controllable models based on complementary learning, such as Rumi-GAN, have been introduced. Rumi-GANs leverage negative examples to enhance the generator's ability to learn positive examples. We evaluate the performance of two controllable GAN variants, CGAN and Rumi-GAN, in generating game levels targeting specific constraints of interest: playability and controllability. This evaluation is conducted under two scenarios: with and without the inclusion of negative examples. The goal is to determine whether incorporating negative examples helps the GAN models avoid generating undesirable outputs. Our findings highlight the strengths and weaknesses of each method in enforcing the generation of specific conditions when generating outputs based on given positive and negative examples.