A New Approach to Plan-Space Explanation: Analyzing Plan-Property Dependencies in Oversubscription Planning

Abstract

In many usage scenarios of AI Planning technology, users will want not just a plan π but an explanation of the space of possible plans, justifying π. In particular, in oversubscription planning where not all goals can be achieved, users may ask why a conjunction A of goals is not achieved by π. We propose to answer this kind of question with the goal conjunctions B excluded by A, i. e., that could not be achieved if A were to be enforced. We formalize this approach in terms of plan-property dependencies, where plan properties are propositional formulas over the goals achieved by a plan, and dependencies are entailment relations in plan space. We focus on entailment relations of the form ∧g∈A g ⇒ ⌝ ∧g∈B g, and devise analysis techniques globally identifying all such relations, or locally identifying the implications of a single given plan property (user question) ∧g∈A g. We show how, via compilation, one can analyze dependencies between a richer form of plan properties, specifying formulas over action subsets touched by the plan. We run comprehensive experiments on adapted IPC benchmarks, and find that the suggested analyses are reasonably feasible at the global level, and become significantly more effective at the local level.

Explanation Generation for Multi-Modal Multi-Agent Path Finding with Optimal Resource Utilization using Answer Set Programming

“…Note that all these questions can be handled by QP, QW, QC and QU queries in the context of mMAPF. Recently, Eifler et al (2020) has presented a method to generate "contrastive" explanations to questions of the last two forms: " Why not p?" where p is a propositional formula describing an plan property.…”

In Defence of Design Patterns for AI Planning Knowledge Models

“…The idea being that an AI system should be able to explain, to some extent, its behaviour to stakeholders. Focusing on the narrower topic dubbed XAIP, as for EXplainable AI Planning, a number of recent works considered the problem of defining what "explainable" means for an automated planning system, explaining and describing generated plans, bridging the gap between machines and human stakeholders, and designing approaches to explain the behaviour of planning systems [6,5,3,22,21,2].…”

The Emerging Landscape of Explainable Automated Planning & Decision Making

“…The explanation is then to identify an exemplary plan or policy that satisfies those constraints thus demonstrating how the computed plan is better. Authors in [20], on the other hand, expect the user queries to be expressed in terms of plan properties which are user-defined binary properties that apply to all valid plans for the problem. The explanation then takes the form of other plan properties that are entailed by those properties.…”

“…Among these works, [41; 19; 37; 20] aim for minimal explanations as a means of selection. Moreover, [37] and [20] could be considered social as they at least specifically try to frame explanations in human understandable terms. Q 4 : "Why is Π not solvable?…”

“…The method discussed in [39] also allows for causal link explanations for abstract tasks, such as in HTN planning [63]. The use of plan properties by [20], interpretable state representations in [81], and subsets of state factors in [41] are more examples of the use of abstraction schemes to simplify the explanation process. Certificates.…”