Goal Agnostic Planning using Maximum Likelihood Paths in Hypergraph World Models
This work addresses planning challenges in AI for autonomous agents, though it appears incremental as it combines classical ML and traditional AI approaches.
The paper tackles the problem of goal-agnostic automated planning for autonomous learning agents by developing a hypergraph-based algorithm and planning method, proving optimal solutions and bounding learning performance with mathematical models.
In this paper, we present a hypergraph--based machine learning algorithm, a datastructure--driven maintenance method, and a planning algorithm based on a probabilistic application of Dijkstra's algorithm. Together, these form a goal agnostic automated planning engine for an autonomous learning agent which incorporates beneficial properties of both classical Machine Learning and traditional Artificial Intelligence. We prove that the algorithm determines optimal solutions within the problem space, mathematically bound learning performance, and supply a mathematical model analyzing system state progression through time yielding explicit predictions for learning curves, goal achievement rates, and response to abstractions and uncertainty. To validate performance, we exhibit results from applying the agent to three archetypal planning problems, including composite hierarchical domains, and highlight empirical findings which illustrate properties elucidated in the analysis.