This AI Is Learning To Create INTERESTING Games

In this video, we cover how a large language model can modify existing games to create novel, interesting game experiences This video is in collaboration with NYU’s Game Innovation Lab and covers work by Graham Todd and collaborators. Link to this week’s paper- GAVEL: Generating Games Via Evolution and Language Models by Todd et al. https://arxiv.org/pdf/2407.09388 Other References: Conflict, Arousal, and Curiosity by Daniel Berlyne: https://psycnet.apa.org/doiLanding?doi=10.1037%2F11164-000 Jurgen Schmidhuber’s work on Interestingness, Curiosity, and Creativity: https://arxiv.org/abs/0709.0674 The Animated Monte-Carlo Tree Search (MCTS) by Thomas Kurbiel: https://medium.com/data-science/the-animated-monte-carlo-tree-search-mcts-c05bb48b018c Evolutionary Game Design by Cameron Browne and Frederic Maire: https://www.researchgate.net/publication/224111054_Evolutionary_Game_Design