Heuristic Search Optimisation Using Planning and Curriculum Learning Techniques

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F23%3A00373634" target="_blank" >RIV/68407700:21230/23:00373634 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1007/978-3-031-49008-8_39" target="_blank" >https://doi.org/10.1007/978-3-031-49008-8_39</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-031-49008-8_39" target="_blank" >10.1007/978-3-031-49008-8_39</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Heuristic Search Optimisation Using Planning and Curriculum Learning Techniques

Popis výsledku v původním jazyce

Learning a well-informed heuristic function for hard planning domains is an elusive problem. Although there are known neural network architectures to represent such heuristic knowledge, it is not obvious what concrete information is learned and whether techniques aimed at understanding the structure help in improving the quality of the heuristics. This paper presents a network model that learns a heuristic function capable of relating distant parts of the state space via optimal plan imitation using the attention mechanism which drastically improves the learning of a good heuristic function. The learning of this heuristic function is further improved by the use of curriculum learning, where newly solved problem instances are added to the training set, which, in turn, helps to solve problems of higher complexities and train from harder problem instances. The methodologies used in this paper far exceed the performances of all existing baselines including known deep learning approaches and classical planning heuristics. We demonstrate its effectiveness and success on grid-type PDDL domains, namely Sokoban, maze-with-teleports and sliding tile puzzles.

Název v anglickém jazyce

Heuristic Search Optimisation Using Planning and Curriculum Learning Techniques

Popis výsledku anglicky

Learning a well-informed heuristic function for hard planning domains is an elusive problem. Although there are known neural network architectures to represent such heuristic knowledge, it is not obvious what concrete information is learned and whether techniques aimed at understanding the structure help in improving the quality of the heuristics. This paper presents a network model that learns a heuristic function capable of relating distant parts of the state space via optimal plan imitation using the attention mechanism which drastically improves the learning of a good heuristic function. The learning of this heuristic function is further improved by the use of curriculum learning, where newly solved problem instances are added to the training set, which, in turn, helps to solve problems of higher complexities and train from harder problem instances. The methodologies used in this paper far exceed the performances of all existing baselines including known deep learning approaches and classical planning heuristics. We demonstrate its effectiveness and success on grid-type PDDL domains, namely Sokoban, maze-with-teleports and sliding tile puzzles.

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

Výsledek vznikl pri realizaci vícero projektů. Více informací v záložce Projekty.

Návaznosti

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

Rok uplatnění

2023

Kód důvěrnosti údajů

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

Název statě ve sborníku

Progress in Artificial Intelligence, 22nd EPIA Conference on Artificial Intelligence, EPIA 2023, Faial Island, Azores, September 5–8, 2023, Proceedings, Part I

ISBN

978-3-031-49007-1

ISSN

0302-9743

e-ISSN

1611-3349

Počet stran výsledku

13

Strana od-do

495-507

Název nakladatele

Springer, Cham

Místo vydání

—

Místo konání akce

Faial Island

Datum konání akce

5. 9. 2023

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

001160573500039