Utilizing Generative Adversarial Networks for Stable Structure Generation in Angry Birds

TL;DR

This paper explores using GANs to generate stable, complex structures in Angry Birds, advancing level design by applying deep learning to physics-based game environments.

Contribution

It introduces a novel encoding method and applies state-of-the-art GAN architectures to generate stable structures for Angry Birds levels.

Findings

GANs can generate diverse stable structures

The encoding method effectively translates level descriptions

Generated structures are playable and physically stable

Abstract

This paper investigates the suitability of using Generative Adversarial Networks (GANs) to generate stable structures for the physics-based puzzle game Angry Birds. While previous applications of GANs for level generation have been mostly limited to tile-based representations, this paper explores their suitability for creating stable structures made from multiple smaller blocks. This includes a detailed encoding/decoding process for converting between Angry Birds level descriptions and a suitable grid-based representation, as well as utilizing state-of-the-art GAN architectures and training methods to produce new structure designs. Our results show that GANs can be successfully applied to generate a varied range of complex and stable Angry Birds structures.