This repository contains the source code to reproduce the experiments in the paper "Lifting Factor Graphs with Some Unknown Factors for New Individuals" by Malte Luttermann, Ralf Möller, and Marcel Gehrke (IJAR 2025).
The code features the LIFAGU algorithm that has initially been presented in the paper "Lifting Factor Graphs with Some Unknown Factors" by Malte Luttermann, Ralf Möller, and Marcel Gehrke (ECSQARU 2023). The initial source code of LIFAGU can be found here. The code in this repository slightly adapts the experimental setup, however, the LIFAGU algorithm itself remains unchanged.
Our implementation uses the Julia programming language.
All experiments were conducted using Julia version 1.11.2 together with the following packages:
- BayesNets v3.4.1
- BenchmarkTools v1.6.0
- Combinatorics v1.0.2
- Graphs v1.9.0
- Multisets v0.4.5
Moreover, we use openjdk version 11.0.20 to run the (lifted) inference
algorithms, which are integrated via
instances/ljt-v1.0-jar-with-dependencies.jar.
Run julia instance_generator.jl in the directory src/ to generate all input
instances for evaluation.
The generated instances are written to the directory instances/ in binary format.
After generating the instances, the experiments can be started by executing
julia run_eval.jl in the directory src/.
All results are written to the directory results/.
To obtain the plots, first run julia prepare_plot.jl in the directory results/
to combine the run times into averages and afterwards run the R script plot.r,
which creates .tex files (also in the directory results/) containing the
plots in TikZ.