mlsanityData to 14 Oct 2026

Simulation-Driven Balancing of Competitive Game Levels with Reinforcement Learning

arXiv 2503.18748 ↗First posted 24 Mar 2025cs.LGcited by 0

On the map: Deep reinforcement learning methods → RL for games and combinatorial optimization

Abstract (arXiv)

The balancing process for game levels in competitive two-player contexts involves a lot of manual work and testing, particularly for non-symmetrical game levels. In this work, we frame game balancing as a procedural content generation task and propose an architecture for automatically balancing of tile-based levels within the PCGRL framework (procedural content generation via reinforcement learning). Our architecture is divided into three parts: (1) a level generator, (2) a balancing agent, and (3) a reward modeling simulation. Through repeated simulations, the balancing agent receives rewards for adjusting the level towards a given balancing objective, such as equal win rates for all players. To this end, we propose new swap-based representations to improve the robustness of playability, thereby enabling agents to balance game levels more effectively and quickly compared to traditional PCGRL. By analyzing the agent's swapping behavior, we can infer which tile types have the most impact on the balance. We validate our approach in the Neural MMO (NMMO) environment in a competitive two-player scenario. In this extended conference paper, we present improved results, explore the applicability of the method to various forms of balancing beyond equal balancing, compare the performance to another search-based approach, and discuss the application of existing fairness metrics to game balancing.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RL for games and combinatorial optimization, one of 15 directions in Deep reinforcement learning methods. Builds on 23 of its 50 references that are in the map. First posted on arXiv 24 Mar 2025.

Builds on

23 of 50 references.

References · 23

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2014Certifying and removing disparate impactcited by 715
  3. 2016Fair prediction with disparate impact: A study of bias in recidivism prediction instrumentscited by 672
  4. 2018Aequitas: A Bias and Fairness Audit Toolkitcited by 147
  5. 2017Procedural Content Generation via Machine Learning (PCGML)cited by 137
  6. 2020PCGRL: Procedural Content Generation via Reinforcement Learningcited by 90
  7. 2020Deep Learning for Procedural Content Generationcited by 73
  8. 2019Neural MMO: A Massively Multiagent Game Environment for Training and Evaluating Intelligent Agentscited by 44
  9. 2021Learning Controllable Content Generatorscited by 40
  10. 2021Adversarial Reinforcement Learning for Procedural Content Generationcited by 36
  11. 2020TOAD-GAN: Coherent Style Level Generation from a Single Examplecited by 22
  12. 2022Learning Controllable 3D Level Generatorscited by 19
11 more
  1. 2020Learning to Generate Levels From Nothingcited by 17
  2. 2018DOOM Level Generation using Generative Adversarial Networkscited by 15
  3. 2022Path of Destruction: Learning an Iterative Level Generator Using a Small Datasetcited by 14
  4. 2022Mutation Models: Learning to Generate Levels by Imitating Evolutioncited by 9
  5. 2019Evolving the Hearthstone Metacited by 8
  6. 2022TropeTwist: Trope-based Narrative Structure Generationcited by 8
  7. 2016Demonstrating the Feasibility of Automatic Game Balancingcited by 5
  8. 2022Generating Real-Time Strategy Game Units Using Search-Based Procedural Content Generation and Monte Carlo Tree Searchcited by 5
  9. 2023Learning of Generalizable and Interpretable Knowledge in Grid-Based Reinforcement Learning Environmentscited by 5
  10. 2023Balancing of competitive two-player Game Levels with Reinforcement Learningcited by 4
  11. 2024GEEvo: Game Economy Generation and Balancing with Evolutionary Algorithmscited by 4

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Level the Level: Balancing Game Levels for Asymmetric Player Archetypes With Reinforcement Learning
  2. 2024PCGRL+: Scaling, Control and Generalization in Reinforcement Learning Level Generators
  3. 2026RuleSmith: Multi-Agent LLMs for Automated Game Balancing
  4. 2024A Framework for Predicting the Impact of Game Balance Changes through Meta Discovery
  5. 2025UBCL: A Reinforcement Learning Framework for Controllable and Diverse Player Behaviors
  6. 2025Video Game Level Design as a Multi-Agent Reinforcement Learning Problem
  7. 2024FightLadder: A Benchmark for Competitive Multi-Agent Reinforcement Learning
  8. 2025Beyond Playtesting: A Generative Multi-Agent Simulation System for Massively Multiplayer Online Games