Playing Text-Adventure Games with Graph-Based Deep Reinforcement Learning

Abstract

Text-based adventure games provide a platform on which to explore reinforcement learning in the context of a combinatorial action space, such as natural language. We present a deep reinforcement learning architecture that represents the game state as a knowledge graph which is learned during exploration. This graph is used to prune the action space, enabling more efficient exploration. The question of which action to take can be reduced to a question-answering task, a form of transfer learning that pre-trains certain parts of our architecture. In experiments using the TextWorld framework, we show that our proposed technique can learn a control policy faster than baseline alternatives. We have also open-sourced our code at https://github.com/rajammanabrolu/KG-DQN.

Proceedings of NAACL-HLT 2019

References in corpus (3)

• Sequence to Sequence Learning with Neural Networks
• The Natural Language Decathlon: Multitask Learning as Question Answering
• TextWorld: A Learning Environment for Text-based Games

Cited by in corpus (7)

• NAIL: A General Interactive Fiction Agent
• Building Dynamic Knowledge Graphs from Text-based Games
• Jointly-Learned State-Action Embedding for Efficient Reinforcement Learning
• Bootstrapped Q-learning with Context Relevant Observation Pruning to Generalize in Text-based Games
• Sentiment Analysis for Reinforcement Learning
• Playing optical tweezers with deep reinforcement learning: in virtual, physical and augmented environments
• Learning Natural Language Generation from Scratch