# Hybrid Encoding For Generating Large Scale Game Level Patterns With Local Variations

arXiv: https://arxiv.org/abs/2105.12960 · HTML: https://mlsanity.com/paper/2105.12960/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 39 references that are in the map.
- First posted on arXiv 27 May 2021.

## Builds on (most cited first, 14 of 14)

- [Large Scale GAN Training for High Fidelity Natural Image Synthesis](https://mlsanity.com/paper/1809.11096/) (2018)
- [HyperNetworks](https://mlsanity.com/paper/1609.09106/) (2016)
- [Illuminating search spaces by mapping elites](https://mlsanity.com/paper/1504.04909/) (2015)
- [Evolving Mario Levels in the Latent Space of a Deep Convolutional Generative Adversarial Network](https://mlsanity.com/paper/1805.00728/) (2018)
- [COCO-GAN: Generation by Parts via Conditional Coordinating](https://mlsanity.com/paper/1904.00284/) (2019)
- [The VGLC: The Video Game Level Corpus](https://mlsanity.com/paper/1606.07487/) (2016)
- [Bootstrapping Conditional GANs for Video Game Level Generation](https://mlsanity.com/paper/1910.01603/) (2019)
- [TOAD-GAN: Coherent Style Level Generation from a Single Example](https://mlsanity.com/paper/2008.01531/) (2020)
- [CPPN2GAN: Combining Compositional Pattern Producing Networks and GANs for Large-scale Pattern Generation](https://mlsanity.com/paper/2004.01703/) (2020)
- [Sequential Segment-based Level Generation and Blending using Variational Autoencoders](https://mlsanity.com/paper/2007.08746/) (2020)
- [Generative Adversarial Network Rooms in Generative Graph Grammar Dungeons for The Legend of Zelda](https://mlsanity.com/paper/2001.05065/) (2020)
- [Convolution by Evolution: Differentiable Pattern Producing Networks](https://mlsanity.com/paper/1606.02580/) (2016)
- [DeepMasterPrints: Generating MasterPrints for Dictionary Attacks via Latent Variable Evolution](https://mlsanity.com/paper/1705.07386/) (2017)
- [Capturing Local and Global Patterns in Procedural Content Generation via Machine Learning](https://mlsanity.com/paper/2005.12579/) (2020)

## Built on by (most cited first, 4 of 4)

- [Evolutionary Reinforcement Learning: A Survey](https://mlsanity.com/paper/2303.04150/) (2023)
- [Measuring Diversity of Game Scenarios](https://mlsanity.com/paper/2404.15192/) (2024)
- [Evolutionary Machine Learning and Games](https://mlsanity.com/paper/2311.16172/) (2023)
- [Text-to-Level Diffusion Models With Various Text Encoders for Super Mario Bros](https://mlsanity.com/paper/2507.00184/) (2025)

## Abstract (arXiv)

Generative Adversarial Networks (GANs) are a powerful indirect genotype-to-phenotype mapping for evolutionary search. Much previous work applying GANs to level generation focuses on fixed-size segments combined into a whole level, but individual segments may not fit together cohesively. In contrast, segments in human designed levels are often repeated, directly or with variation, and organized into patterns (the symmetric eagle in Level 1 of The Legend of Zelda, or repeated pipe motifs in Super Mario Bros). Such patterns can be produced with Compositional Pattern Producing Networks (CPPNs). CPPNs define latent vector GAN inputs as a function of geometry, organizing segments output by a GAN into complete levels. However, collections of latent vectors can also be evolved directly, producing more chaotic levels. We propose a hybrid approach that evolves CPPNs first, but allows latent vectors to evolve later, combining the benefits of both approaches. These approaches are evaluated in Super Mario Bros. and The Legend of Zelda. We previously demonstrated via a Quality-Diversity algorithm that CPPNs better cover the space of possible levels than directly evolved levels. Here, we show that the hybrid approach (1) covers areas that neither of the other methods can, and (2) achieves comparable or superior QD scores.