Handling Overlaps When Lifting Gaussian Bayesian Networks

Hartwig, Mattis; Braun, Tanya; Möller, Ralf

doi:10.24963/ijcai.2021/581

Handling Overlaps When Lifting Gaussian Bayesian Networks

Mattis Hartwig, Tanya Braun, Ralf Möller

Download BibTeX Cite

Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence. Main Track. Pages 4228-4234.

https://doi.org/10.24963/ijcai.2021/581

Keywords

Uncertainty in AI: Bayesian Networks
Uncertainty in AI: Exact Probabilistic Inference
Uncertainty in AI: Statistical Relational AI
Uncertainty in AI: Graphical Models

Abstract

Gaussian Bayesian networks are widely used for modeling the behavior of continuous random variables. Lifting exploits symmetries when dealing with large numbers of isomorphic random variables. It provides a more compact representation for more efficient query answering by encoding the symmetries using logical variables. This paper improves on an existing lifted representation of the joint distribution represented by a Gaussian Bayesian network (lifted joint), allowing overlaps between the logical variables. Handling overlaps without grounding a model is critical for modelling real-world scenarios.

Specifically, this paper contributes (i) a lifted joint that allows overlaps in logical variables and (ii) a lifted query answering algorithm using the lifted joint. Complexity analyses and experimental results show that - despite overlaps - constructing a lifted joint and answering queries on the lifted joint outperform their grounded counterparts significantly.

Mattis Hartwig, Tanya Braun, and Ralf Möller. 2021. Handling Overlaps When Lifting Gaussian Bayesian Networks. In Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21. IJCAI, 4228–4234. https://doi.org/10.24963/ijcai.2021/581

@inproceedings{ijcai2021p581,
  title     = {Handling Overlaps When Lifting Gaussian Bayesian Networks},
  author    = {Hartwig, Mattis and Braun, Tanya and Möller, Ralf},
  booktitle = {Proceedings of the Thirtieth International Joint Conference on
               Artificial Intelligence, {IJCAI-21}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Prof. Zhi-Hua Zhou},
  pages     = {4228--4234},
  year      = {2021},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2021/581},
  url       = {https://doi.org/10.24963/ijcai.2021/581},
}

TY  - CPAPER
TI  - Handling Overlaps When Lifting Gaussian Bayesian Networks
AU  - Hartwig, Mattis
AU  - Braun, Tanya
AU  - Möller, Ralf
T2  - Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2021
SP  - 4228
EP  - 4234
DO  - 10.24963/ijcai.2021/581
UR  - https://doi.org/10.24963/ijcai.2021/581
ER  -