TextWorld
Microsoft TextWorld is an open-source, extensible engine that both generates and simulates text games. You can use it to train reinforcement learning (RL) agents to learn skills such as language understanding and grounding, combined with sequential decision making.
You are navigating through a house. You've just entered a serious study. There is a gross looking mantle in the room. It has nothing on it. You see a closed rusty toolbox. Now why would someone leave that there?Looks like there is a locked door. Find the key to unlock the door. You should try going east.
Unlock the door to text-based adventure (opens in new tab)
Text-based games may seem primitive next to the beautifully rendered graphics of today, but to succeed at even the simplest such game, humans must use a special set of skills. We seamlessly and simultaneously comprehend language descriptions, plan our next moves, visualize the environment, remember important information, and generalize past experiences to new situations. AI agents don’t yet possess such capabilities, but they are the key to general intelligence. We can help an AI agent to learn these skills.
Try TextWorld today
The default environment is a house where the RL agent can learn basic domestic skills such as putting objects in containers, opening and closing doors, eating food, and more.
You can integrate your agent in the TextWorld framework, then train your agent to solve a potentially limitless number of multi-step quests in a potentially limitless number of text-based worlds. To learn more and get the code, visit our GitHub page.
Questions?
Send your questions and feedback to [email protected].