Category A1; Team name: MappingComplexityLab; Dataset: HypBench

[robertjankowski]

Checklist

• [x] My pull request has a clear and explanatory title.
• [x] My pull request passes the Linting test.
• [x] I added appropriate unit tests and I made sure the code passes all unit tests. (refer to comment below)
• [x] My PR follows PEP8 guidelines. (refer to comment below)
• [x] My code is properly documented, using numpy docs conventions, and I made sure the documentation renders properly.
• [ ] I linked to issues and PRs that are relevant to this PR.

Description

This is an implementation of HypBench [1] for TAG-DS Topological Deep Learning Challenge 2025: Expanding the Data Landscape.

HypBench is a comprehensive benchmarking framework for graph machine learning, focusing on the performance of GNNs across varied network structures. Utilizing the geometric soft configuration model in hyperbolic space, one can generate synthetic networks with realistic topological properties and node feature vectors. This approach enables us to assess the impact of network properties, such as topology-feature correlation, degree distributions, local density of triangles (or clustering), and homophily, on the effectiveness of different GNN architectures.

HypBench relies on the $\mathbb{S}^1/\mathbb{H}^2$ model [2,3] to generate synthetic networks and on the bipartite- $\mathbb{S}^1/\mathbb{H}^2$ model for generating synthetic node features [4] (represented as a bipartite graph).

Time complexity is $\mathcal{O}(N^2 + N M)$ where $N$ is the number of nodes in the network and $M$ is the number of features.

In addition to the HypBenchDatasetLoader class, I implemented a tutorials/tutorial_hypbench_dataset.ipynb where one can visualize this geometric network model, and observe the impact of each model's parameters.

[1] Aliakbarisani, R., Jankowski, R., Serrano, M., & Boguñá, M. (2024). Hyperbolic Benchmarking Unveils Network Topology-Feature Relationship in GNN Performance. arXiv preprint arXiv:2406.02772. [2] Serrano, M. Á., Krioukov, D., & Boguná, M. (2008). Self-similarity of complex networks and hidden metric spaces. Physical Review Letters, 100(7), 078701. [3] Krioukov, D., Papadopoulos, F., Kitsak, M., Vahdat, A., & Boguná, M. (2010). Hyperbolic geometry of complex networks. Physical Review E—Statistical, Nonlinear, and Soft Matter Physics, 82(3), 036106. [4] Aliakbarisani, R., Serrano, M. Á., & Boguñá, M. (2025). Feature-enriched hyperbolic network geometry. Physical Review Research, 7(3), 033036.

Issue

Additional context

[review-notebook-app[bot]]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB