Fast Algorithms for Hypergraph PageRank with Applications to Semi-Supervised Learning
Authors: Konstantinos Ameranis, Adela Frances Depavia, Lorenzo Orecchia, Erasmo Tani
ICML 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In addition to giving strong theoretical guarantees, we empirically showcase the speed of our algorithms on benchmark instances of semi-supervised learning on categorical data. |
| Researcher Affiliation | Academia | 1Department of Computer Science, University of Chicago, Chicago, USA 2Computational and Applied Mathematics, University of Chicago, Chicago, USA. Correspondence to: Lorenzo Orecchia <orecchia@uchicago.edu>. |
| Pseudocode | Yes | Algorithm 1 with prox-generating function 1/2 2 R for min F(x) s, x in Theorem 4.1 |
| Open Source Code | Yes | Code for conducting all experiments in Appendix C and Appendix D is publically available on Git Hub.3 ... 3Full link address: https://github.com/Orecchia-Research-Group/hypergraph_diffusions |
| Open Datasets | Yes | As a benchmark, we adopt the semi-supervised multi-class classification task on a small set of UCI datasets (Kelly et al.) which was considered in previous works on the topic (Zhou et al., 2003; Hein et al., 2013; Li et al., 2020). |
| Dataset Splits | No | The paper describes using a small subset of labeled points for semi-supervised learning, but it does not provide specific training, validation, or test dataset splits (e.g., percentages or counts) for reproducibility. |
| Hardware Specification | Yes | All experiments were run on a server with a 24core Intel Xeon Silver 4116 CPU @ 2.10GHz processor and 128gb RAM. |
| Software Dependencies | No | The paper mentions C++ and MATLAB implementations but does not specify version numbers for these or any other software libraries or dependencies. |
| Experiment Setup | Yes | We use teleportation constant α = 0.5. For the hypergraph, we use Algorithm 1 with 50 iterations to compute the PPR vector... We chose this single value of λ as it proved representative of the relative behavior of each method. |