Medijski sadržaj
- Tweetovi
- Tweetovi i odgovori
- Medijski sadržaj, trenutna stranica.
-
Here's our team, in case you were wondering. We're all standing up straight, or at least perpendicular to some high-dimensional hyperplane that is not depicted because of the inherent limitations of projecting onto a 2D plane.pic.twitter.com/K6drseaiJr
Prikaži ovu nit -
Seriously, though. Here's Sutton and Barto (1998), from the introduction to the book Reinforcement Learning. Too many don't understand this.pic.twitter.com/vLGr7v1jmE
Prikaži ovu nit -
It is well known from "non-deep" agent learning research (say, evolutionary robotics work and game AI work from early to mid 2000s) that the sensor representation is extremely important to the agent's ability to learn. http://julian.togelius.com/VanHoorn2009Hierarchical.pdf …pic.twitter.com/Ugj9IDsdDR
Prikaži ovu nit -
In our new paper, "Rotation, Translation, and Cropping for Zero-Shot Generalization" by
@yooceii@Amidos2006@FilipoGiovanni and I show that the agent-centric perspective is better in the sense that the agent learns policies that generalize better. https://arxiv.org/abs/2001.09908 pic.twitter.com/q7jLjBRjLL
Prikaži ovu nit -
To the left, you see a trained agent playing a level of a game. To the right, you see the same playthrough from an agent-centric perspective: cropped, translated, and rotated with the agent in the center. Which perspective is the best input for the agent? https://arxiv.org/abs/2001.09908 pic.twitter.com/7bCtBp8xUG
Prikaži ovu nit -
A really nice piece from
@nyutandon about the NYU Game Innovation Lab@NYUGameLab and the research we do here. Featuring@Amidos2006@FilipoGiovanni@ruben_torrado@Bumblebor and myself. https://engineering.nyu.edu/news/virtuous-circle-ai-and-games-game-innovation-lab … Thanks@KarlPGreenberg for the write-up!pic.twitter.com/RKS2i0Fol9
-
The difference, instead, is that PCGRL doesn't search for levels. Instead, it searches for level generators. It learns to generate levels. So it moves time consumption from inference time to training time.pic.twitter.com/VFfGXhjLGN
Prikaži ovu nit -
You may wonder how PCGRL relates to search-based PCG (SBPCG). It's not the fact that one approach and the other something. It would be completely possible to do PCGRL with neuroevolution. In this paper we use PPO, but the choice of RL algorithm is not important.pic.twitter.com/AYjYBfpmVz
Prikaži ovu nit -
Finally, the Wide representation differs drastically, in the sense that the network can choose to edit any tile at any timestep. This creates a much larger timestep, but generally much shorter episodes.pic.twitter.com/fiW6zPKhgv
Prikaži ovu nit -
What PCGRL does is that it learns to at any point take the action that maximizes expected future quality. In other words, it learns to improve the level one step at a time. This is particularly interesting for interactive and mixed-initiative approaches.pic.twitter.com/IVnBwJHtWw
Prikaži ovu nit -
The reward function differs between the game. In the maze scenario (previous tweet) it's simply how long the longest path is. In Sokoban, shows here, we want to have as many boxes as possible, as long as the level is solvable. So we include a simple solver in the reward function.pic.twitter.com/2LYRq0Ka5K
Prikaži ovu nit -
In the Turtle representation, inspired by Turtle graphics, the neural network can move around freely and edit where it wants. The "agent" has the option to change the tile where it "stands", or move in any cardinal direction. This representation is perhaps the most gamelike one.pic.twitter.com/SGbAklbhl8
Prikaži ovu nit -
In all cases, we initialize level generation with random levels. This is important to ensure that the network does not simply learn to produce a single level, but instead learns to produce a level generator that can produce an unlimited amount of good new levels.pic.twitter.com/4guKfOgVYN
Prikaži ovu nit -
We train these networks by presenting a sequence of random positions, but we usually present positions to the network sequentially, like scan lines on an old TV. Interestingly, the network works with almost any position order.pic.twitter.com/rHGwag4wdo
Prikaži ovu nit -
In the narrow representation, the neural network cannot control where it edits the map; it is presented with a series of positions, and for each of them it needs to decide whether to swap out a particular tile or not.pic.twitter.com/oBuS796rq9
Prikaži ovu nit -
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 pic.twitter.com/J9c8O5EIM7
Prikaži ovu nit -
我很高兴地向大家宣布:我和
@Yannakakis 共同撰写的《人工智能与游戏》一书已经被译为中文并出版了!谢谢@_JialinLiu 的帮助!pic.twitter.com/pSD48veM0u
Prikaži ovu nit -
In case I wanted to press any of these buttons (I do not), which would be the more hygienic choice to press?pic.twitter.com/UT2je8LvJt – mjesto: Amsterdam Airport Schiphol (AMS)
-
NB: I'm not writing a novel. I'm barely writing anything at all right now. Here's a road in Skåne in December.pic.twitter.com/RIfu3jgjdP
Prikaži ovu nit -
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.