Automated learning of hierarchical task networks for controlling minecraft agents
Chanh Nguyen, Noah Reifsnyder, Sriram Gopalakrishnan, Héctor Muñoz‐Avila · 2017
In this paper, we present an agent that learns Hierarchical Task Network (HTN) knowledge from observing a player performing actions in Minecraft. From these observations, the agent learns the tasks that the player pursues and how to achieve these tasks. We present an HTN learning algorithm and report on experiments of the agent assisting a player performing tasks in Minecraft.