Weisfeiler and Leman go Machine Learning: The Story so far
Authors: Christopher Morris, Yaron Lipman, Haggai Maron, Bastian Rieck, Nils M. Kriege, Martin Grohe, Matthias Fey, Karsten Borgwardt
JMLR 2023 | Venue PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Theoretical | Here, we give a comprehensive overview of the algorithm’s use in a machine-learning setting, focusing on the supervised regime. We discuss the theoretical background, show how to use it for supervised graph and node representation learning, discuss recent extensions, and outline the algorithm’s connection to (permutation-)equivariant neural architectures. Moreover, we give an overview of current applications and future directions to stimulate further research. |
| Researcher Affiliation | Collaboration | Christopher Morris EMAIL Department of Computer Science RWTH Aachen University Aachen, Germany Yaron Lipman EMAIL Meta AI Research Department of Computer Science and Applied Mathematics Weizmann Institute of Science Rehovot, Israel Haggai Maron EMAIL NVIDIA Research Tel Aviv, Israel Bastian Rieck EMAIL AIDOS Lab, Institute of AI for Health Helmholtz Zentrum M unchen and Technical University of Munich Munich, Germany Nils M. Kriege EMAIL Faculty of Computer Science, University of Vienna, Vienna, Austria Research Network Data Science, University of Vienna, Vienna, Austria Martin Grohe EMAIL Department of Computer Science RWTH Aachen University Aachen, Germany Matthias Fey EMAIL Kumo.AI Mountain View, CA Karsten Borgwardt EMAIL Machine Learning & Computational Biology Lab Department of Biosystems Science and Engineering ETH Z urich, Basel, Switzerland |
| Pseudocode | No | The paper describes algorithms such as the Weisfeiler Leman algorithm and GNN architectures using mathematical equations and textual descriptions, but does not include any explicitly labeled pseudocode or algorithm blocks. |
| Open Source Code | No | The paper mentions existing open-source and commercial software libraries such as RDKit and Open Eye Graph Sim TK (Section 1.2.2) that implement certain functionalities. However, it does not provide any statement or link for the source code of the methodology described in *this* survey paper. |
| Open Datasets | Yes | For the comparison of machine learning methods, several standard benchmark data sets were introduced (Hu et al., 2020; Morris et al., 2020a), which contain graphs representing various objects and concepts such as small molecules, proteins, and protein-protein interactions, citation networks, as well as letters, fingerprints, and cuneiform signs. |
| Dataset Splits | No | This paper is a survey and does not present its own experimental results. Therefore, it does not specify any training/test/validation dataset splits. |
| Hardware Specification | No | This paper is a survey and does not present its own experimental results. Therefore, it does not specify any hardware used for running experiments. |
| Software Dependencies | No | The paper is a survey and primarily discusses existing methodologies. It does not describe new software requiring specific dependencies with version numbers for its own work. While it mentions RDKit and Open Eye Graph Sim TK as tools, these are not dependencies for the paper's own methodology. |
| Experiment Setup | No | This paper is a survey and does not present its own experimental results. Therefore, it does not describe a specific experimental setup or hyperparameter values. |