Mixed-Initiative Level Design with RL Brush
Not on the map. Why?
Abstract (arXiv)
This paper introduces RL Brush, a level-editing tool for tile-based games designed for mixed-initiative co-creation. The tool uses reinforcement-learning-based models to augment manual human level-design through the addition of AI-generated suggestions. Here, we apply RL Brush to designing levels for the classic puzzle game Sokoban. We put the tool online and tested it in 39 different sessions. The results show that users using the AI suggestions stay around longer and their created levels on average are more playable and more complex than without.
Key facts. Cited by 13 papers in the map (data through 7 Oct 2026). First posted on arXiv 6 Aug 2020.
Builds on
No reference list yet.
Built on by
Citing papers per year: 2021: 3; 2022: 6; 2023: 1; 2024: 1; 2025: 2.
13 papers.
Sort:
- 2021Learning Controllable Content Generatorscited by 40
- 2021Illuminating Diverse Neural Cellular Automata for Level Generationcited by 35
- 2022Learning Controllable 3D Level Generatorscited by 19
- 2022Path of Destruction: Learning an Iterative Level Generator Using a Small Datasetcited by 14
- 2022Mutation Models: Learning to Generate Levels by Imitating Evolutioncited by 9
- 2022Generative Personas That Behave and Experience Like Humanscited by 7
- 2024PCGRL+: Scaling, Control and Generalization in Reinforcement Learning Level Generatorscited by 7
- 2022Diversity-based Deep Reinforcement Learning Towards Multidimensional Difficulty for Fighting Game AIcited by 3
- 2022Start Small: Training Controllable Game Level Generators without Training Data by Learning at Multiple Sizescited by 2
- 2025Human-Aligned Procedural Level Generation Reinforcement Learning via Text-Level-Sketch Shared Representationcited by 2
- 2021Toward Co-creative Dungeon Generation via Transfer Learningcited by 1
- 2023Lode Enhancer: Level Co-creation Through Scalingcited by 1
1 more
- 2025Video Game Level Design as a Multi-Agent Reinforcement Learning Problem6 Oct 2025
- 2025Human-Aligned Procedural Level Generation Reinforcement Learning via Text-Level-Sketch Shared Representation13 Aug 2025
- 2024PCGRL+: Scaling, Control and Generalization in Reinforcement Learning Level Generators22 Aug 2024
- 2023Lode Enhancer: Level Co-creation Through Scaling3 Aug 2023
- 2022Diversity-based Deep Reinforcement Learning Towards Multidimensional Difficulty for Fighting Game AI4 Nov 2022
- 2022Start Small: Training Controllable Game Level Generators without Training Data by Learning at Multiple Sizes29 Sep 2022
- 2022Generative Personas That Behave and Experience Like Humans26 Aug 2022
- 2022Learning Controllable 3D Level Generators27 Jun 2022
- 2022Mutation Models: Learning to Generate Levels by Imitating Evolution11 Jun 2022
- 2022Path of Destruction: Learning an Iterative Level Generator Using a Small Dataset21 Feb 2022
- 2021Illuminating Diverse Neural Cellular Automata for Level Generation12 Sep 2021
- 2021Toward Co-creative Dungeon Generation via Transfer Learning27 Jul 2021
1 more
- 2021Learning Controllable Content Generators6 May 2021
- 2021Learning Controllable Content Generatorscited by 40
- 2021Illuminating Diverse Neural Cellular Automata for Level Generationcited by 35
- 2022Learning Controllable 3D Level Generatorscited by 19
- 2022Path of Destruction: Learning an Iterative Level Generator Using a Small Datasetcited by 14
- 2022Mutation Models: Learning to Generate Levels by Imitating Evolutioncited by 9
- 2022Generative Personas That Behave and Experience Like Humanscited by 7
- 2024PCGRL+: Scaling, Control and Generalization in Reinforcement Learning Level Generatorscited by 7
- 2022Diversity-based Deep Reinforcement Learning Towards Multidimensional Difficulty for Fighting Game AIcited by 3
- 2022Start Small: Training Controllable Game Level Generators without Training Data by Learning at Multiple Sizescited by 2
- 2025Human-Aligned Procedural Level Generation Reinforcement Learning via Text-Level-Sketch Shared Representationcited by 2
- 2021Toward Co-creative Dungeon Generation via Transfer Learningcited by 1
- 2023Lode Enhancer: Level Co-creation Through Scalingcited by 1
12 most cited of 13 All 13 (JSON)
- 2025Video Game Level Design as a Multi-Agent Reinforcement Learning Problem6 Oct 2025
- 2025Human-Aligned Procedural Level Generation Reinforcement Learning via Text-Level-Sketch Shared Representation13 Aug 2025
- 2024PCGRL+: Scaling, Control and Generalization in Reinforcement Learning Level Generators22 Aug 2024
- 2023Lode Enhancer: Level Co-creation Through Scaling3 Aug 2023
- 2022Diversity-based Deep Reinforcement Learning Towards Multidimensional Difficulty for Fighting Game AI4 Nov 2022
- 2022Start Small: Training Controllable Game Level Generators without Training Data by Learning at Multiple Sizes29 Sep 2022
- 2022Generative Personas That Behave and Experience Like Humans26 Aug 2022
- 2022Learning Controllable 3D Level Generators27 Jun 2022
- 2022Mutation Models: Learning to Generate Levels by Imitating Evolution11 Jun 2022
- 2022Path of Destruction: Learning an Iterative Level Generator Using a Small Dataset21 Feb 2022
- 2021Illuminating Diverse Neural Cellular Automata for Level Generation12 Sep 2021
- 2021Toward Co-creative Dungeon Generation via Transfer Learning27 Jul 2021
12 newest of 13 All 13 (JSON)