Symmetry-Invariant Novelty Heuristics via Unsupervised Weisfeiler-Leman Features
This addresses a specific issue in planning and heuristic search for AI researchers, but it appears incremental as it builds on existing WLFs for novelty detection.
The paper tackled the problem of novelty heuristics in heuristic search not being symmetry invariant, which can cause redundant exploration, by proposing to use Weisfeiler-Leman Features (WLFs) for synthesizing lifted, domain-independent novelty heuristics that are invariant to symmetric states, with experiments on benchmark suites yielding promising results.
Novelty heuristics aid heuristic search by exploring states that exhibit novel atoms. However, novelty heuristics are not symmetry invariant and hence may sometimes lead to redundant exploration. In this preliminary report, we propose to use Weisfeiler-Leman Features for planning (WLFs) in place of atoms for detecting novelty. WLFs are recently introduced features for learning domain-dependent heuristics for generalised planning problems. We explore an unsupervised usage of WLFs for synthesising lifted, domain-independent novelty heuristics that are invariant to symmetric states. Experiments on the classical International Planning Competition and Hard To Ground benchmark suites yield promising results for novelty heuristics synthesised from WLFs.