<div class="csl-bib-body">
<div class="csl-entry">Maskey, S., Parviz, A., Thiessen, M., Stärk, H., Sadikaj, Y., & Maron, H. (2022, December 3). <i>Generalized Laplacian Positional Encoding for Graph Representation Learning</i> [Poster Presentation]. NeurIPS 2022 Workshop on Symmetry and Geometry in Neural Representations, New Orleans, United States of America (the). https://doi.org/10.34726/3908</div>
</div>
-
dc.identifier.uri
http://hdl.handle.net/20.500.12708/176001
-
dc.identifier.uri
https://doi.org/10.34726/3908
-
dc.description.abstract
Graph neural networks (GNNs) are the primary tool for processing graph-structured data. Unfortunately, the most commonly used GNNs, called Message Passing Neural Networks (MPNNs) suffer from several fundamental limitations. To overcome these limitations, recent works have adapted the idea of positional encodings to graph data. This paper draws inspiration from the recent success of Laplacian-based positional encoding and defines a novel family of positional encoding schemes for graphs. We accomplish this by generalizing the optimization problem that defines the Laplace embedding to more general dissimilarity functions rather than the 2-norm used in the original formulation. This family of positional encodings is then instantiated by considering p-norms. We discuss a method for calculating these positional encoding schemes, implement it in PyTorch and demonstrate how the resulting positional encoding captures different properties of the graph. Furthermore, we demonstrate that this novel family of positional encodings can improve the expressive power of MPNNs. Lastly, we present preliminary experimental results.
en
dc.language.iso
en
-
dc.rights.uri
http://creativecommons.org/licenses/by/4.0/
-
dc.subject
Machine Learning
en
dc.subject
Graph Neural Networks
en
dc.subject
Positional Encoding
en
dc.subject
Graph Laplacian
en
dc.title
Generalized Laplacian Positional Encoding for Graph Representation Learning
en
dc.type
Presentation
en
dc.type
Vortrag
de
dc.rights.license
Creative Commons Namensnennung 4.0 International
de
dc.rights.license
Creative Commons Attribution 4.0 International
en
dc.identifier.doi
10.34726/3908
-
dc.contributor.affiliation
Ludwig-Maximilians-University of Munich, Germany
-
dc.contributor.affiliation
New Jersey Institute of Technology, USA
-
dc.contributor.affiliation
Massachusetts Institute of Technology (MIT), MA, USA