Papers of the day   All papers

PCGRL: Procedural Content Generation via Reinforcement Learning

Comments

Julian Togelius: What if we see level generation as a game, where each action changes the level in some way? Well, we could use RL to learn to "play" level generation! We introduce Procedural Content Generation via Reinforcement Learning (PCGRL), a new paradigm for PCG. https://arxiv.org/abs/2001.09212 https://t.co/J9c8O5EIM7

8 replies, 273 likes


hardmaru: Procedural Content Generation via Reinforcement Learning “A new approach to procedural content generation in games, where level design is framed as a game (as a sequential task problem), and the content generator itself is learned.” https://arxiv.org/abs/2001.09212 https://t.co/2B6P6tlh9d

2 replies, 215 likes


Ahmed Khalifa: Just in case if you missed last night's tweet, Our work about #PCGRL is out. Go download the system and play around (it is built on #OpenAIGYM) github link: https://github.com/amidos2006/gym-pcgrl paper link: https://arxiv.org/abs/2001.09212 #PCG #AI #RL

3 replies, 27 likes


Julian Togelius: @hardmaru More info - and gifs! - here: https://twitter.com/togelius/status/1222025962604441600?s=19

0 replies, 26 likes


Julian Togelius: Our new paper, by @Amidos2006 @FilipoGiovanni @Smearle_RH and myself, lays out the conceptual framework for learning level generators with reinforcement learning and provides an initial Deep RL implementation. https://arxiv.org/abs/2001.09212

1 replies, 17 likes


Philip Bontrager: Our new paper on training computer agents to design video game levels using reinforcement learning

0 replies, 10 likes


roadrunner01: PCGRL: Procedural Content Generation via Reinforcement Learning pdf: https://arxiv.org/pdf/2001.09212.pdf abs: https://arxiv.org/abs/2001.09212 https://t.co/xXGA1AHofQ

0 replies, 10 likes


arxiv: PCGRL: Procedural Content Generation via Reinforcement Learning. http://arxiv.org/abs/2001.09212 https://t.co/bAWg5Dps9D

0 replies, 7 likes


Sam Earle: We're working on a framework for Procedural Content Generation using Reinforcement Learning: teaching agents to play the "game" of game-design. Cute and bizarre level design-styles abound. Many possible applications; many ways we might expand the system. Code is available!

0 replies, 7 likes


Michael Green: Really cool work out by my friend and colleagues at @NYUGameLab involving #PCGRL! You can play with it and read the paper at the links below!

0 replies, 1 likes


Content

Found on Jan 28 2020 at https://arxiv.org/pdf/2001.09212.pdf

PDF content of a computer science paper: PCGRL: Procedural Content Generation via Reinforcement Learning