Answer Set Programming (ASP) is a well-established declarative AI formalism for knowledge representation and reasoning. ASP systems were successfully applied to both industrial and academic problems. Nonetheless, their performance can be improved by embedding domain-specific heuristics into their solving process. However, the development of domain-specific heuristics often requires both a deep knowledge of the domain at hand and a good understanding of the fundamental working principles of the ASP solvers. In this paper, we investigate the use of deep learning techniques to automatically generate domain-specific heuristics for ASP solvers targeting the well-known graph coloring problem. Empirical results show that the idea is promising: the performance of the ASP solver wasp can be improved.

Deep Learning for the Generation of Heuristics in Answer Set Programming: A Case Study of Graph Coloring

Dodaro C.;Ilardi D.;Oneto L.;
2022-01-01

Abstract

Answer Set Programming (ASP) is a well-established declarative AI formalism for knowledge representation and reasoning. ASP systems were successfully applied to both industrial and academic problems. Nonetheless, their performance can be improved by embedding domain-specific heuristics into their solving process. However, the development of domain-specific heuristics often requires both a deep knowledge of the domain at hand and a good understanding of the fundamental working principles of the ASP solvers. In this paper, we investigate the use of deep learning techniques to automatically generate domain-specific heuristics for ASP solvers targeting the well-known graph coloring problem. Empirical results show that the idea is promising: the performance of the ASP solver wasp can be improved.
978-3-031-15706-6
978-3-031-15707-3
File in questo prodotto:
File Dimensione Formato  
C111.pdf

accesso aperto

Descrizione: Contributo in atti di convegno
Tipologia: Documento in versione editoriale
Dimensione 458.32 kB
Formato Adobe PDF
458.32 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11567/1102736
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 0
social impact