AILOJun 22, 2021

Knowing How to Plan

arXiv:2106.11504v1
Originality Incremental advance
AI Analysis

This work addresses planning challenges in multi-agent systems with epistemic reasoning, though it appears incremental as it builds on existing know-how logics.

The paper tackles the problem of higher-order epistemic planning with know-how goals, such as ensuring a condition while preventing adversaries from knowing how to undermine it, by developing a PTIME algorithm for model checking over finite epistemic transition systems.

Various planning-based know-how logics have been studied in the recent literature. In this paper, we use such a logic to do know-how-based planning via model checking. In particular, we can handle the higher-order epistemic planning involving know-how formulas as the goal, e.g., find a plan to make sure p such that the adversary does not know how to make p false in the future. We give a PTIME algorithm for the model checking problem over finite epistemic transition systems and axiomatize the logic under the assumption of perfect recall.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes