Game Description Generation (GDG) is the task of generating a game description written in a Game Description Language (GDL) from natural language text. Previous studies have explored generation methods leveraging the contextual understanding capabilities of Large Language Models (LLMs); however, accurately reproducing the game features of the game descriptions remains a challenge. In this paper, we propose reinforcement learning-based fine-tuning of LLMs for GDG (RLGDG). Our training method simultaneously improves grammatical correctness and fidelity to game concepts by introducing both grammar rewards and concept rewards. Furthermore, we adopt a two-stage training strategy where Reinforcement Learning (RL) is applied following Supervised Fine-Tuning (SFT). Experimental results demonstrate that our proposed method significantly outperforms baseline methods using SFT alone.

Many games often share common ideas or aspects between them, such as their rules, controls, or playing area. However, in the context of General Game Playing (GGP) for board games, this area remains under-explored. We propose to formalise the notion of "game concept", inspired by terms generally used by game players and designers. Through the Ludii General Game System, we describe concepts for several levels of abstraction, such as the game itself, the moves played, or the states reached. This new GGP feature associated with the ludeme representation of games opens many new lines of research. The creation of a hyper-agent selector, the transfer of AI learning between games, or explaining AI techniques using game terms, can all be facilitated by the use of game concepts. Other applications which can benefit from game concepts are also discussed, such as the generation of plausible reconstructed rules for incomplete ancient games, or the implementation of a board game recommender system.

Michael Cook, a 30-year-old senior research fellow at the University of Falmouth, has built an AI capable of imagining new video games from scratch. Cook calls the machine Angelina, a recursive acronym that stands for "A Novel Game-Evolving Labrat I've Named Angelina" (a joke that Cook says got old pretty quickly). Since its earliest form, in 2011, it has created hundreds of experimental video games, received acclaim in an international game-making competition, and had its work featured in a New York gallery exhibit. Game-making algorithms are almost as old as video games, but their use has typically been limited to generating terrain and other simple digital art. The next frontier is using increasingly sophisticated machine-learning techniques to design entirely new kinds of games that have, to date, evaded the human imagination.