Ollivier-Ricci Curvature for Hypergraphs: A Unified Framework

Authors: Corinna Coupette, Sebastian Dalleiger, Bastian Rieck

ICLR 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Through extensive experiments on synthetic and real-world hypergraphs from different domains, we demonstrate that ORCHID curvatures are both scalable and useful to perform a variety of hypergraph tasks in practice.
Researcher Affiliation Academia Corinna Coupette1, Sebastian Dalleiger1,2, Bastian Rieck3,4 1Max Planck Institute for Informatics 2CISPA Helmholtz Center for Information Security 3AIDOS Lab, Institute of AI for Health, Helmholtz Munich 4Technical University of Munich (TUM)
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks. It describes mathematical definitions and proofs but no algorithm steps formatted like code.
Open Source Code Yes We discuss our implementation and results in more detail in Appendices A.4 and A.5, and make all our code, data, and results publicly available.1 (...) 1https://doi.org/10.5281/zenodo.7624573
Open Datasets Yes DBLP: All DBLP data is released under a CC0 license and freely available in one XML file that is updated regularly. We obtained the XML dump dated September 1, 2022 from https://dblp.org/xml/release/. (...) NDC: We downloaded the NDC data from https://download.open.fda.gov/ drug/ndc/drug-ndc-0001-of-0001.json.zip on August 21, 2022.
Dataset Splits No The paper does not provide specific dataset split information (exact percentages, sample counts, citations to predefined splits, or detailed splitting methodology) needed to reproduce the data partitioning for training, validation, or testing.
Hardware Specification Yes Our experiments are run on AMD EPYC 7702 CPUs with up to 256 cores.
Software Dependencies No The paper states: "We implement ORCHID in Julia and Python." However, it does not specify version numbers for these languages or any specific libraries/solvers used, which is required for reproducible software dependencies.
Experiment Setup No The paper discusses the parameters of its proposed framework (α, µ, AGG) and how they impact curvatures. It also mentions using spectral clustering and RBF/Wasserstein kernels. However, it does not provide concrete hyperparameter values for general experimental setups, such as learning rates, batch sizes, or optimizer settings, nor a dedicated section for detailed experimental configuration.