Partial Evaluation in Junction Trees
Aug 31, 2022·
,,,·
0 min read
Martin Roa Villescas
Patrick wijnings
Sander stuijk
Henk corporaal
Abstract
One prominent method to perform inference on probabilistic graphical models is the probability propagation in trees of clusters (PPTC) algorithm. In this paper, we demonstrate the use of partial evaluation, an established technique from the compiler domain, to improve the performance of online Bayesian inference using the PPTC algorithm in the context of observed evidence. We present a metaprogramming-based method to transform a base program into an optimized version by precomputing the static input at compile time while guaranteeing behavioral equivalence. We achieve an inference time reduction of 21% on average for the Promedas benchmark.
Publication
In 2022 25th Euromicro Conference on Digital System Design (DSD)
Junction Trees
Partial Evaluation
Bayesian Inference
Probabilistic Graphical Models-
Message Passing
Authors
Teacher & Researcher
Martin Roa Villescas holds a BSc in Electronic Engineering from the National
University of Colombia and an MSc in Embedded Systems from Eindhoven
University of Technology (TU/e). He worked at Philips Research as an
embedded software designer from 2013 to 2018. He later returned to TU/e for
his doctoral research in model-based machine learning, carried out within
the PhD-Teaching Assistant trajectory combining research and teaching. Since
2023, he has been working at Fontys University of Applied Sciences in the
Netherlands, where he teaches in the Information and Communication
Technology program and conducts research in robotics and smart industry.
Authors
Authors
Authors