Summary of Gavel: Generating Games Via Evolution and Language Models, by Graham Todd et al.
GAVEL: Generating Games Via Evolution and Language Models
by Graham Todd, Alexander Padula, Matthew Stephenson, Éric Piette, Dennis J.N.J. Soemers, Julian Togelius
First submitted to arxiv on: 12 Jul 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper explores automatic game generation, focusing on the Ludii game description language, which represents over 1000 board games. The authors draw inspiration from large language models and evolutionary computation to train a model that generates novel and interesting games by intelligently mutating and recombining rules and mechanics expressed as code. The approach demonstrates both quantitative and qualitative results, showing its ability to generate new and innovative games in regions of the potential rules space not covered by existing Ludii dataset games. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Automated game generation is a tough task! The challenge is to create new and interesting games while making sure they follow certain rules. Researchers have tried before, but mostly with limited rule representations and special tricks for specific types of games. This paper takes a different approach by using a big language model and some clever evolutionary techniques to generate games in the Ludii game description language. The result is a program that can create new and exciting games, including ones that don’t exist in the dataset! You can even play some of these generated games online. |
Keywords
» Artificial intelligence » Language model
