Super Mario as a String: Platformer Level Generation Via LSTMs

Abstract

The procedural generation of video game levels has existed for at least 30 years, but only recently have machine learning approaches been used to generate levels without specifying the rules for generation. A number of these have looked at platformer levels as a sequence of characters and performed generation using Markov chains. In this paper we examine the use of Long Short-Term Memory recurrent neural networks (LSTMs) for the purpose of generating levels trained from a corpus of Super Mario Brothers levels. We analyze a number of different data representations and how the generated levels fit into the space of human authored Super Mario Brothers levels.

References in corpus (6)

• Sequence to Sequence Learning with Neural Networks
• Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks
• Going Deeper with Convolutions
• DRAW: A Recurrent Neural Network For Image Generation
• Learning to Execute
• Toward Game Level Generation from Gameplay Videos

Cited by in corpus (14)

• Deep Learning for Procedural Content Generation
• Friend, Collaborator, Student, Manager: How Design of an AI-Driven Game Level Editor Affects Creators
• The VGLC: The Video Game Level Corpus
• Level Generation Through Large Language Models
• Efficient Generation of Structured Objects with Constrained Adversarial Networks
• Conceptual Game Expansion
• Pairing Character Classes in a Deathmatch Shooter Game via a Deep-Learning Surrogate Model
• Generating Lode Runner Levels by Learning Player Paths with LSTMs
• Explainable PCGML via Game Design Patterns
• On the Importance of Environments in Human-Robot Coordination
• Co-Creative Level Design via Machine Learning
• Conditional Level Generation and Game Blending
• Explainability via Responsibility
• Tile Embedding: A General Representation for Procedural Level Generation via Machine Learning