Saved in:
Bibliographic Details
Main Authors: Luttermann, Malte, Möller, Ralf, Gehrke, Marcel
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.04089
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912311215652864
author Luttermann, Malte
Möller, Ralf
Gehrke, Marcel
author_facet Luttermann, Malte
Möller, Ralf
Gehrke, Marcel
contents Lifting exploits symmetries in probabilistic graphical models by using a representative for indistinguishable objects, allowing to carry out query answering more efficiently while maintaining exact answers. In this paper, we investigate how lifting enables us to perform probabilistic inference for factor graphs containing unknown factors, i.e., factors whose underlying function of potential mappings is unknown. We present the Lifting Factor Graphs with Some Unknown Factors (LIFAGU) algorithm to identify indistinguishable subgraphs in a factor graph containing unknown factors, thereby enabling the transfer of known potentials to unknown potentials to ensure a well-defined semantics of the model and allow for (lifted) probabilistic inference. We further extend LIFAGU to incorporate additional background knowledge about groups of factors belonging to the same individual object. By incorporating such background knowledge, LIFAGU is able to further reduce the ambiguity of possible transfers of known potentials to unknown potentials.
format Preprint
id arxiv_https___arxiv_org_abs_2504_04089
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Lifting Factor Graphs with Some Unknown Factors for New Individuals
Luttermann, Malte
Möller, Ralf
Gehrke, Marcel
Artificial Intelligence
Machine Learning
Lifting exploits symmetries in probabilistic graphical models by using a representative for indistinguishable objects, allowing to carry out query answering more efficiently while maintaining exact answers. In this paper, we investigate how lifting enables us to perform probabilistic inference for factor graphs containing unknown factors, i.e., factors whose underlying function of potential mappings is unknown. We present the Lifting Factor Graphs with Some Unknown Factors (LIFAGU) algorithm to identify indistinguishable subgraphs in a factor graph containing unknown factors, thereby enabling the transfer of known potentials to unknown potentials to ensure a well-defined semantics of the model and allow for (lifted) probabilistic inference. We further extend LIFAGU to incorporate additional background knowledge about groups of factors belonging to the same individual object. By incorporating such background knowledge, LIFAGU is able to further reduce the ambiguity of possible transfers of known potentials to unknown potentials.
title Lifting Factor Graphs with Some Unknown Factors for New Individuals
topic Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2504.04089