Tanya Braun, M. Sc.

- Research Assistant -

 

Institut für Informationssysteme
Universität zu Lübeck
Ratzeburger Allee 160 ( Gebäude 64 - 2.OG )
D-23562 Lübeck

Telefon: +49 451 3101 5708 
Fax : +49 451 3101 5704
email : Öffnet ein Fenster zum Versenden einer E-Mail E-Mail for Tanya Braun
 
 

Curriculum Vitae

Research Interests

  • Machine Learning
  • Probabilistic Graphical Models
  • Lifted Inference

For my PhD, I work on probabilistic first-order formalisms where the domain objects are known. In these formalisms, the standard approach for inference with first-order constructs include lifted variable elimination (LVE) for single queries. To handle multiple queries efficiently, the junction tree algorithm takes advantage of the underlying knowledge base being constant. To benefit from these advantages in the first-order setting, we transfer the idea of lifting to the junction tree algorithm and introduce the lifted junction tree algorithm (LJT). It aims at reducing computations by introducing a first-order cluster representation of a knowledge base, called first-order junction trees, which compactly represents symmetries, and using LVE in its computations.

Given multiple queries, e.g., in machine learning applications, our approach enables us to compute answers faster than existing approaches tailored for single queries and the propositional version of the junction tree algorithm.

 

Publications

2018

  • Marcel Gehrke, Tanya Braun, Ralf Möller: Lifted Dynamic Junction Tree Algorithm
    to be published in: Proceedings of the International Conference on Conceptual Structures, 2018
    BibTeX
  • Tanya Braun, Ralf Möller: Lifted Most Probable Explanation
    to be published in: Proceedings of the International Conference on Conceptual Structures, 2018, Springer
    BibTeX
  • Tanya Braun, Ralf Möller: Counting and Conjunctive Queries in the Lifted Junction Tree Algorithm - Extended Version
    in: Postproceedings of the 5th International Workshop on Graph Structures for Knowledge Representation and Reasoning, GKR 2017, Melbourne, Australia, August 21, 2017, 2018, Springer
    BibTeX
to top

2017

  • Tanya Braun, Felix Kuhr, Ralf Möller: Unsupervised Text Annotations
    in: Proceedings of the 6th Workshop on Dynamics of Knowledge and Belief (DKB-2017) and the 5th Workshop KI & Kognition (KIK-2017) co-located with 40th German Conference on Artificial Intelligence (KI 2017),, 2017, 25.-29.09., CEUR Workshop Proceedings, Vol.1928, p.23-30
    Website BibTeX
  • Tanya Braun, Ralf Möller: Preventing Groundings and Handling Evidence in the Lifted Junction Tree Algorithm
    in: KI 2017: Advances in Artificial Intelligence. KI 2017 - 40th Annual German Conference on AI, Dortmund, Germany, September 25-29, 2017, 2017, Springer, LNCS, Vol.10505, p.85-98
    DOI BibTeX
  • Tanya Braun, Ralf Möller: Counting and Conjunctive Queries in the Lifted Junction Tree Algorithm
    in: Graph Structures for Knowledge Representation and Reasoning - 5th International Workshop (GKR 2017), Melbourne, Australia, 2017, 21. August
    BibTeX
to top

2016

  • Tanya Braun, Ralf Möller: Lifted Junction Tree Algorithm
    in: KI 2016: Advances in Artificial Intelligence - 39th Annual German Conference on AI, Klagenfurt, Austria, September 26-30, 2016, 2016, Gerhard Friedrich, Malte Helmert, Franz Wotawa (Ed.), Springer, Lecture Notes in Computer Science, Vol.9904, p.30-42
    DOI BibTeX
  • Tanya Braun, Ralf Möller: Lifted Junction Tree Algorithm
    IFIS, Universität zu Lübeck, 2016, Long version of the KI 2016 conference paper
    BibTeX
to top