Planning with Dynamically Estimated Action Costs

TL;DR

This paper introduces a flexible planning approach that dynamically chooses between multiple action cost estimators to balance computational efficiency and estimation accuracy, improving scalability and reliability in AI planning.

Contribution

It generalizes deterministic planning with multiple cost estimators and develops a new $A^*$-based algorithm with theoretical guarantees and practical efficiency.

Findings

Significant runtime savings over existing methods

Effective balancing of computation time and estimation uncertainty

Enhanced scalability for large planning problems

Abstract

Information about action costs is critical for real-world AI planning applications. Rather than rely solely on declarative action models, recent approaches also use black-box external action cost estimators, often learned from data, that are applied during the planning phase. These, however, can be computationally expensive, and produce uncertain values. In this paper we suggest a generalization of deterministic planning with action costs that allows selecting between multiple estimators for action cost, to balance computation time against bounded estimation uncertainty. This enables a much richer -- and correspondingly more realistic -- problem representation. Importantly, it allows planners to bound plan accuracy, thereby increasing reliability, while reducing unnecessary computational burden, which is critical for scaling to large problems. We introduce a search algorithm, generalizing $A^*$, that solves such planning problems, and additional algorithmic extensions. In addition to theoretical guarantees, extensive experiments show considerable savings in runtime compared to alternatives.