|
RockIt: Exploiting Parallelism and Symmetry for MAP Inference in Statistical Relational Models
Noessner, Jan
;
Niepert, Mathias
;
Stuckenschmidt, Heiner
|
URL:
|
http://www.aaai.org/ocs/index.php/AAAI/AAAI13/pape...
|
|
Additional URL:
|
http://publications.wim.uni-mannheim.de/informatik...
|
|
Document Type:
|
Conference or workshop publication
|
|
Year of publication:
|
2013
|
|
Book title:
|
Proceedings of the Twenty-Seventh AAAI Conference on Artificial Intelligence, July 14-18, 2012, Bellevue, Washington, USA
|
|
Page range:
|
739-745
|
|
Date of the conference:
|
July 14-18, 2013
|
|
Publisher:
|
DesJardins, Marie
|
|
Place of publication:
|
Menlo Park, Calif.
|
|
Publishing house:
|
AAAI Press
|
|
ISBN:
|
978-1-57735-615-8
|
|
Publication language:
|
English
|
|
Institution:
|
School of Business Informatics and Mathematics > Practical Computer Science II: Artificial Intelligence (Stuckenschmidt 2009-)
|
|
Subject:
|
004 Computer science, internet
|
|
Keywords (English):
|
Markov Logic , Statistical Relational Models , Lifted Inference , Parallelization
|
|
Abstract:
|
ROCKIT is a maximum a-posteriori (MAP) query engine for statistical relational models. MAP inference in graphical models is an optimization problem which can be compiled to integer linear programs (ILPs). We describe several advances in translating MAP queries
to ILP instances and present the novel meta-algorithm cutting plane aggregation (CPA). CPA exploits local context-specific symmetries and bundles up sets of linear constraints. The resulting counting constraints lead
to more compact ILPs and make the symmetry of
the ground model more explicit to state-of-the-art ILP solvers. Moreover, ROCKIT parallelizes most parts of the MAP inference pipeline taking advantage of ubiquitous shared-memory multi-core architectures.
We report on extensive experiments with Markov logic network (MLN) benchmarks showing that ROCKIT outperforms the state-of-the-art systems ALCHEMY, MARKOV THEBEAST, and TUFFY both in terms of efficiency and quality of results.
|
 | Dieser Eintrag ist Teil der Universitätsbibliographie. |
 Search Authors in
BASE:
Noessner, Jan
;
Niepert, Mathias
;
Stuckenschmidt, Heiner
Google Scholar:
Noessner, Jan
;
Niepert, Mathias
;
Stuckenschmidt, Heiner
ORCID:
Noessner, Jan, Niepert, Mathias and Stuckenschmidt, Heiner  ORCID: https://orcid.org/0000-0002-0209-3859
You have found an error? Please let us know about your desired correction here: E-Mail
Actions (login required)
 |
Show item |
|
|
