Option Discovery for Autonomous Generation of Symbolic Knowledge
This addresses the challenge of autonomous knowledge generation for AI agents, though it appears incremental as it builds on existing domains and methods.
The paper tackles the problem of enabling an artificial agent to autonomously explore an environment without predefined goals, discovering and learning options that can be abstracted into a symbolic planning model. The result demonstrates that these discovered options allow the agent to generate symbolic plans to achieve extrinsic goals, tested empirically in the Treasure Game domain.
In this work we present an empirical study where we demonstrate the possibility of developing an artificial agent that is capable to autonomously explore an experimental scenario. During the exploration, the agent is able to discover and learn interesting options allowing to interact with the environment without any pre-assigned goal, then abstract and re-use the acquired knowledge to solve possible tasks assigned ex-post. We test the system in the so-called Treasure Game domain described in the recent literature and we empirically demonstrate that the discovered options can be abstracted in an probabilistic symbolic planning model (using the PPDDL language), which allowed the agent to generate symbolic plans to achieve extrinsic goals.