Preventing Groundings and Handling Evidence in the Lifted Junction Tree Algorithm

Tanya Braun, Ralf Möller

Abstract

For inference in probabilistic formalisms with first-order constructs, lifted variable elimination (LVE) is one of the standard approaches for single queries. To handle multiple queries efficiently, the lifted junction tree algorithm (LJT) uses a specific representation of a first-order knowledge base and LVE in its computations. Unfortunately, LJT induces unnecessary groundings in cases where the standard LVE algorithm, GC-FOVE, has a fully lifted run. Additionally, LJT does not handle evidence explicitly. We extend LJT (i) to identify and prevent unnecessary groundings and (ii) to effectively handle evidence in a lifted manner. Given multiple queries, e.g., in machine learning applications, our extension computes answers faster than LJT and GC-FOVE.
OriginalspracheEnglisch
TitelKI 2017: Advances in Artificial Intelligence
Redakteure/-innenGabriele Kern-Isberner, Johannes Fürnkranz, Matthias Thimm
Seitenumfang14
Band10505
ErscheinungsortCham
Herausgeber (Verlag)Springer International Publishing
Erscheinungsdatum19.09.2017
Seiten85-98
ISBN (Print)978-3-319-67189-5
ISBN (elektronisch)978-3-319-67190-1
DOIs
PublikationsstatusVeröffentlicht - 19.09.2017
Veranstaltung40th Annual German Conference on Artificial Intelligence - Dortmund, Deutschland
Dauer: 25.09.201729.09.2017
Konferenznummer: 199309

DFG-Fachsystematik

  • 409-01 Theoretische Informatik

Fingerprint

Untersuchen Sie die Forschungsthemen von „Preventing Groundings and Handling Evidence in the Lifted Junction Tree Algorithm“. Zusammen bilden sie einen einzigartigen Fingerprint.

Zitieren