NAIL: A General Interactive Fiction Agent

Abstract

Interactive Fiction (IF) games are complex textual decision making problems. This paper introduces NAIL, an autonomous agent for general parser-based IF games. NAIL won the 2018 Text Adventure AI Competition, where it was evaluated on twenty unseen games. This paper describes the architecture, development, and insights underpinning NAIL's performance.

References in corpus (2)

• Skip-Thought Vectors
• Using reinforcement learning to learn how to play text-based games