Answering Hindsight Queries with Lifted Dynamic Junction Trees

Marcel Gehrke, Tanya Braun, Ralf Möller

Abstract

The lifted dynamic junction tree algorithm (LDJT) efficiently answers filtering and prediction queries for probabilistic relational temporal models by building and then reusing a first-order cluster representation of a knowledge base for multiple queries and time steps. We extend LDJT to (i) solve the smoothing inference problem to answer hindsight queries by introducing an efficient backward pass and (ii) discuss different options to instantiate a first-order cluster representation during a backward pass. Further, our relational forward backward algorithm makes hindsight queries to the very beginning feasible. LDJT answers multiple temporal queries faster than the static lifted junction tree algorithm on an unrolled model, which performs smoothing during message passing.
Original languageEnglish
Number of pages8
Publication statusPublished - 02.07.2018
Event27th International Joint Conference on Artificial Intelligence - Stockholm, Sweden
Duration: 13.07.201819.07.2018
Conference number: 140653

Conference

Conference27th International Joint Conference on Artificial Intelligence
Abbreviated title IJCAI 2018
Country/TerritorySweden
CityStockholm
Period13.07.1819.07.18

DFG Research Classification Scheme

  • 4.43-01 Theoretical Computer Science

Fingerprint

Dive into the research topics of 'Answering Hindsight Queries with Lifted Dynamic Junction Trees'. Together they form a unique fingerprint.

Cite this