TextWorld: A Learning Environment for Text-based Games
This provides a tool for researchers in reinforcement learning and natural language processing to study generalization and transfer learning in text-based games, though it is incremental as it builds on existing text game frameworks.
The authors introduced TextWorld, a sandbox learning environment for training and evaluating reinforcement learning agents on text-based games, enabling users to handcraft or automatically generate new games with control over difficulty and language, and they evaluated baseline agents on benchmark games.
We introduce TextWorld, a sandbox learning environment for the training and evaluation of RL agents on text-based games. TextWorld is a Python library that handles interactive play-through of text games, as well as backend functions like state tracking and reward assignment. It comes with a curated list of games whose features and challenges we have analyzed. More significantly, it enables users to handcraft or automatically generate new games. Its generative mechanisms give precise control over the difficulty, scope, and language of constructed games, and can be used to relax challenges inherent to commercial text games like partial observability and sparse rewards. By generating sets of varied but similar games, TextWorld can also be used to study generalization and transfer learning. We cast text-based games in the Reinforcement Learning formalism, use our framework to develop a set of benchmark games, and evaluate several baseline agents on this set and the curated list.