Expressiveness of Graph Neural Networks in Planning Domains

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F24%3A00377976" target="_blank" >RIV/68407700:21230/24:00377976 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1609/icaps.v34i1.31486" target="_blank" >https://doi.org/10.1609/icaps.v34i1.31486</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1609/icaps.v34i1.31486" target="_blank" >10.1609/icaps.v34i1.31486</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Expressiveness of Graph Neural Networks in Planning Domains

Popis výsledku v původním jazyce

Graph Neural Networks (GNNs) have become the standard method of choice for learning with structured data, demonstrating particular promise in classical planning. Their inherent invariance under symmetries of the input graphs endows them with superior generalization capabilities, compared to their symmetry-oblivious counterparts. However, this comes at the cost of limited expressive power. Particularly, GNNs cannot distinguish between graphs that satisfy identical sentences of C2 logic. To leverage GNNs for learning policies in PDDL domains, one needs to encode the contextual representation of the planning states as graphs. The expressiveness of this encoding, coupled with a specific GNN architecture, then hinges on the absence of indistinguishable states necessitating distinct actions. This paper provides a comprehensive theoretical and statistical exploration of such situations in PDDL domains across diverse natural encoding schemes and GNN models.

Název v anglickém jazyce

Expressiveness of Graph Neural Networks in Planning Domains

Popis výsledku anglicky

Graph Neural Networks (GNNs) have become the standard method of choice for learning with structured data, demonstrating particular promise in classical planning. Their inherent invariance under symmetries of the input graphs endows them with superior generalization capabilities, compared to their symmetry-oblivious counterparts. However, this comes at the cost of limited expressive power. Particularly, GNNs cannot distinguish between graphs that satisfy identical sentences of C2 logic. To leverage GNNs for learning policies in PDDL domains, one needs to encode the contextual representation of the planning states as graphs. The expressiveness of this encoding, coupled with a specific GNN architecture, then hinges on the absence of indistinguishable states necessitating distinct actions. This paper provides a comprehensive theoretical and statistical exploration of such situations in PDDL domains across diverse natural encoding schemes and GNN models.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

—

Návaznosti

R - Projekt Ramcoveho programu EK

Rok uplatnění

2024

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Proceedings of the Thirty-Fourth International Conference on Automated Planning and Scheduling

ISBN

978-1-57735-889-3

ISSN

2334-0835

e-ISSN

2334-0843

Počet stran výsledku

9

Strana od-do

281-289

Název nakladatele

AAAI Press

Místo vydání

Menlo Park, California

Místo konání akce

Banff

Datum konání akce

1. 6. 2024

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—