Exploiting Macro-Actions in Learning GPT-Based General Planning Policies

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21730%2F25%3A00387218" target="_blank" >RIV/68407700:21730/25:00387218 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1109/ICTAI66417.2025.00133" target="_blank" >https://doi.org/10.1109/ICTAI66417.2025.00133</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/ICTAI66417.2025.00133" target="_blank" >10.1109/ICTAI66417.2025.00133</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Exploiting Macro-Actions in Learning GPT-Based General Planning Policies

Popis výsledku v původním jazyce

Transformer-based architectures have revolutionized natural language processing and represent a significant promise for advancing policy learning in generalized planning tasks. In particular, PlanGPT demonstrated remarkable results in generating solution plans in different planning domains. For each domain, it is trained on a domain-specific dataset composed of randomly generated planning problems and corresponding solution plans. This work proposes a novel extension to the PlanGPT framework by incorporating macro-actions, high-level actions that encapsulate sequences of primitive actions, which are a well-established technique in classical planning known to improve planning efficiency by guiding exploration through shortcuts. Leveraging this concept, we investigate the impact of macro-actions on the learning process of PlanGPT and whether they mitigate the occurrence of violated preconditions caused by complex object relationships, particularly in domains where these interactions frequently lead to invalid plans due to precondition violations. Experimental results indicate that integrating macroactions improves coverage in several challenging domains and reduces generation time, highlighting the enhanced learning capabilities of PlanGPT when supported by macro-actions.

Název v anglickém jazyce

Exploiting Macro-Actions in Learning GPT-Based General Planning Policies

Popis výsledku anglicky

Transformer-based architectures have revolutionized natural language processing and represent a significant promise for advancing policy learning in generalized planning tasks. In particular, PlanGPT demonstrated remarkable results in generating solution plans in different planning domains. For each domain, it is trained on a domain-specific dataset composed of randomly generated planning problems and corresponding solution plans. This work proposes a novel extension to the PlanGPT framework by incorporating macro-actions, high-level actions that encapsulate sequences of primitive actions, which are a well-established technique in classical planning known to improve planning efficiency by guiding exploration through shortcuts. Leveraging this concept, we investigate the impact of macro-actions on the learning process of PlanGPT and whether they mitigate the occurrence of violated preconditions caused by complex object relationships, particularly in domains where these interactions frequently lead to invalid plans due to precondition violations. Experimental results indicate that integrating macroactions improves coverage in several challenging domains and reduces generation time, highlighting the enhanced learning capabilities of PlanGPT when supported by macro-actions.

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

<a href="/cs/project/GA25-18003S" target="_blank" >GA25-18003S: Použití formálních gramatik a automatů pro získávání a verifikování doménové řídící informace v plánování</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2025

Kód důvěrnosti údajů

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

Název statě ve sborníku

Proceedings of 2025 IEEE 37th International Conference on Tools with Artificial Intelligence ICTAI 2025

ISBN

979-8-3315-4919-0

ISSN

2375-0197

e-ISSN

2375-0197

Počet stran výsledku

5

Strana od-do

915-919

Název nakladatele

IEEE Computer Society

Místo vydání

Los Alamitos

Místo konání akce

Athens

Datum konání akce

3. 11. 2025

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—