Generating Super Mario Bros Levels With Text-Conditional Diffusion Models

This page presents research in Procedural Content Generation via Machine Learning done in collaboration with undergraduate students Olivia Kilday, Bess Hagan, Emilio Salas, and Reid Williams as part of Southwestern University's Summer research program.

Source code for training your own Mario Diffusion models is available on GitHub. The GitHub repository also has simple instructions for downloading our models from Hugging Face and generating your own Mario levels without even training a model!

An arXiv pre-print explaining our approach and results is available here.

Video Demonstrating Our Interactive GUI For Level Creation

Related Publications

Peer-Reviewed Journal Articles

Jacob Schrum, Benjamin Capps, Kirby Steckel, Vanessa Volz, and Sebastian Risi (2023). Hybrid Encoding For Generating Large Scale Game Level Patterns With Local Variations, IEEE Transactions on Games (ToG 2023). March. Volume 15, No. 1, pages 46 - 55. IEEE. (appendix and levels, software).

Peer-Reviewed Conference Publications

Benjamin Capps, and Jacob Schrum (2021). Using Multiple Generative Adversarial Networks to Build Better-Connected Levels for Mega Man, Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2021). (presentation, slides, levels, software).

Jacob Schrum, Vanessa Volz, and Sebastian Risi (2020). CPPN2GAN: Combining Compositional Pattern Producing Networks and GANs for Large-scale Pattern Generation, Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2020). (presentation, slides, software).

Jacob Schrum, Jake Gutierrez, Vanessa Volz, Jialin Liu, Simon M. Lucas, and Sebastian Risi (2020). Interactive Evolution and Exploration Within Latent Level-Design Space of Generative Adversarial Networks, Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2020). (presentation, slides, movie, software).

Jake Gutierrez, and Jacob Schrum (2020). Generative Adversarial Network Rooms in Generative Graph Grammar Dungeons for The Legend of Zelda, Proceedings of the Congress on Evolutionary Computation (CEC 2020). (presentation, slides, movies, software).

Vanessa Volz, Jacob Schrum, Jialin Liu, Simon M. Lucas, Adam M. Smith, and Sebastian Risi (2018). Evolving Mario Levels in the Latent Space of a Deep Convolutional Generative Adversarial Network, Proceedings of the Genetic and Evolutionary Computation Conference (GECCO 2018). (slides, explanation movie, evolution movie, software). Best Paper Award in Digital Entertainment Technologies and Arts

Extended Abstracts

Kirby Steckel, and Jacob Schrum (2021). Illuminating the Space of Beatable Lode Runner Levels Produced By Various Generative Adversarial Networks, Companion to the Proceedings of the Conference on Genetic and Evolutionary Computation (GECCO Companion 2021). (poster, levels, software). (also extended into a technical report)

Technical Reports

Kirby Steckel, and Jacob Schrum (2021). Illuminating the Space of Beatable Lode Runner Levels Produced By Various Generative Adversarial Networks, ArXiv e-print: 2101.07868, January, 2021. (levels, software).

Dagstuhl Reports

Dan Ashlock, Cameron Browne, Simon Colton, Amy K. Hoover, Jialin Liu, Simon M. Lucas, Mark J. Nelson, Diego Perez Liebana, Sebastian Risi, Jacob Schrum, Adam M. Smith, Julian Togelius, and Vanessa Volz (2018). Game Search Space Design and Representation, Dagstuhl Reports: Artificial and Computational Intelligence in Games: AI-Driven Game Design (Dagstuhl Seminar 17471), Volume 7, No. 11, pages 93 - 95. Editors: Pieter Spronck and Elisabeth André and Michael Cook and Mike Preuß. Schloss Dagstuhl--Leibniz-Zentrum fuer Informatik

Undergraduate Poster Presentations Supervised

Maxx Batterton (2021). Evaluating Quality Diversity Success by Transferring MAP Elites Archives, Southwestern University Summer Research Open House. (software).

Benjamin Capps (2020). Evolving Mega Man Levels Using Generative Adversarial Networks, Southwestern University Summer Research Open House. (levels, software).

Kirby Steckel (2020). Evolving Lode Runner Levels with Generative Adversarial Networks, Southwestern University Summer Research Open House. (levels, software).

Jake Gutierrez (2019). Creating Zelda Dungeons with a Graph Grammar and a Generative Adversarial Network, Southwestern University Science Research Open House. (movies, software).

Related Movies and Images

Miscellaneous Content

Summer 2021: Quality Diversity and Creative Divergent Search: SCOPE Research Presentation made by my SCOPE Summer research students to present to other SCOPE students

Fall 2020: Procedural Content Generation for Games with Generative Adversarial Networks: Presentation for the Games AI Research Group at Queen Mary University of London (video)

Fall 2020: Interactively Evolving Video Game Levels with Generative Adversarial Networks: 403 Lecture for Math and CS Department

Fall 2020: Evolving Mega Man Levels with Generative Adversarial Networks: Virtual SCOPE Open House Website

Fall 2020: Evolving Lode Runner Levels with Generative Adversarial Networks: Virtual SCOPE Open House Website

Summer 2020: Generating Video Game Levels Using AI: SCOPE Research Presentation made by my SCOPE Summer research students to present to other SCOPE students

Summer 2019: Playing and Creating Games With Deep Neural Networks: "Mad Science Monday" presentation made by my SCOPE Summer research students to present to other SCOPE students

Fall 2018: Evolutionary Computation Applied to Digital Entertainment and the Arts, poster presented at the President's Appreciation Celebration for Southwestern University donors.

Fall 2018: Evolving Mario Levels in the Latent Space of a Deep Convolutional Generative Adversarial Network: presented to Southwestern University students as a 107 Lecture.

Summer 2018: The machines have taught themselves to make Mario levels, article in Fast Company about my recent GECCO 2018 paper on generating levels for Super Mario Bros.

Spring 2018: Bored with your video game? Artificial intelligence could create new levels on the fly, article in Science about my recent GECCO 2018 paper on generating levels for Super Mario Bros.

Spring 2018: Doom and Super Mario could be a lot tougher now AI is building levels, article in The Register about my recent GECCO 2018 paper on generating levels for Super Mario Bros.

Last Updated: 7/1/2025