AI Paper Atlas

Research map / Graph learning / Graph algorithms and clustering

Learning Heuristic Functions with Graph Neural Networks for Numeric Planning

Valerio Borelli, Alfonso Gerevini, Enrico Scala, Ivan Serina

AAAI 2026 · Graph learning › Graph algorithms and clustering · approach: Neural combinatorial optimization on graphs

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
feed numeric values to the per-condition MLPs on edges (h_n) instead of abstracting numeric conditions as boolean (h_c) (ablation)GNN numeric planning heuristic · IPC 2023 numeric domainstotal coverage: 21 -> 75Improved
add numeric structure (h_c) over ignoring it (h_0) (ablation)GNN numeric planning heuristic · IPC 2023 numeric domainstotal coverage: 11 -> 21Improved
use numeric-value encoding h_n instead of h_c on a domain with few numeric conditions (ablation)GNN numeric planning heuristic · Mprimecoverage: 9 -> 8Hurt
use smooth-max aggregation instead of add aggregation (ablation)GNN numeric planning heuristic · numeric planning domainscoverageHurt
use mean or max aggregation instead of add (ablation)GNN numeric planning heuristic · numeric planning domainscoverageNo clear effect
train on suboptimal plans in addition to optimal plans (ablation)GNN numeric planning heuristic · numeric planning training datavalidation errorHurt
decompose n-ary numeric goals into a chain of n-1 overlapping binary goalsGNN numeric planning heuristic · Fo-Farmlandaverage expanded nodes: 531201 -> 56.30Improved
use learned h_n instead of traditional h_mrp on Fo-Farmlandgreedy best-first search · Fo-Farmlandcoverage: 5 -> 14Improved
use learned h_n instead of traditional h_mrp on Sailinggreedy best-first search · Sailingcoverage: 20 -> 2Hurt
use h_n instead of graph-kernel h_ccWLF_rank on domains with numeric goals and preconditionsgreedy best-first search · Counterscoverage: 3 -> 10Improved
use h_n instead of h_ccWLF_rank on boolean-goal domainsgreedy best-first search · Sailingcoverage: 5 -> 2Hurt
use h_n instead of h_ccWLF_rank on Mprimegreedy best-first search · Mprimecoverage: 13 -> 8Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy