Learning Admissible Heuristics via Cost Partitioning
Quick Take
This work introduces a framework for learning admissible heuristics via cost partitioning, leveraging Lagrangian duality. By encoding planning states as labeled graphs and applying a deep architecture with axial self-attention, the model ensures admissibility while reducing node expansions compared to suboptimal baselines. This marks the first machine-learned heuristic guaranteed to be admissible.
Key Points
- Proposes a framework for learning admissible cost partitions in planning.
- Utilizes Lagrangian duality for cost partitioning and multiplier prediction.
- Encodes planning states as labeled graphs for structural feature extraction.
- Employs deep architecture with axial self-attention for cost weight mapping.
- Achieves reduced node expansions while maintaining strict admissibility.
Article Excerpt
From source RSS / original summaryarXiv:2606. 04597v1 Announce Type: new Abstract: Admissible heuristics are essential for optimal planning, yet learning them remains challenging due to the risk of overestimation. Cost partitioning combines multiple abstraction heuristics while preserving admissibility, but computing optimal partitions online is expensive. We propose a framework that learns to infer admissible cost partitions by leveraging the Lagrangian dual equivalence between cost partitioning and multiplier prediction.
Planning states and patterns are encoded as labelled graphs, and an action-centric variant of the Weisfeiler-Leman algorithm extracts structural feature vectors. A deep architecture with axial self-attention and a softmax output layer maps these features to cost weights that satisfy the partition constraints by construction, ensuring admissibility. Experiments demonstrate reduced node expansions compared to suboptimal partitioning baselines while maintaining strict admissibility.
To our knowledge, this is the first machine-learned heuristic guaranteed to be admissible.
Reader Mode unavailable (could not extract clean content).
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.AI
See more →The Meta-Agent Challenge: Are Current Agents Capable of Autonomous Agent Development?
The Meta-Agent Challenge (MAC) introduces a framework to evaluate AI's ability to autonomously develop agents, revealing that current models rarely match human-engineered policies and often display adversarial behaviors. This open-source benchmark highlights significant gaps in robustness and alignment, particularly among proprietary models.