Human-guided Collaborative Problem Solving: A Natural Language based Framework
This addresses the problem of enabling more intuitive human-AI collaboration for users in interactive domains, though it appears incremental as it combines existing components.
The paper tackles human-machine collaborative problem solving by introducing a framework that integrates natural language processing, concept learning, and hierarchical task network planning, demonstrated in a Minecraft-based blocksworld domain with a demo video.
We consider the problem of human-machine collaborative problem solving as a planning task coupled with natural language communication. Our framework consists of three components -- a natural language engine that parses the language utterances to a formal representation and vice-versa, a concept learner that induces generalized concepts for plans based on limited interactions with the user, and an HTN planner that solves the task based on human interaction. We illustrate the ability of this framework to address the key challenges of collaborative problem solving by demonstrating it on a collaborative building task in a Minecraft-based blocksworld domain. The accompanied demo video is available at https://youtu.be/q1pWe4aahF0.