Citation

Tynchenko, Valeriya V. and Kurashkin, Sergei O. and Borodulin, Aleksei S. and Connie, Tee and Hammoud, Ahmad and Tynchenko, Vadim S. (2026) Is There a Best Hypergraph Neural Network? A Significance-Aware Recomputation and Statistical Audit of DHG-Bench. Machine Learning and Knowledge Extraction, 8 (8). p. 224. ISSN 2504-4990

Text make-08-00224.pdf - Published Version Restricted to Repository staff only Download (763kB)

Abstract

Deep hypergraph learning is evaluated almost entirely through leaderboards that rank methods by mean accuracy over a few random seeds, usually without significance testing. Is there a best hypergraph neural network, or does the apparent ordering reflect seed noise? We independently recomputed the node-classification track of DHG-Bench on a single GPU with twenty random seeds (against five upstream) and a different software stack, and applied a four-layer statistical audit to the per-seed accuracies: a reproducibility check, per-dataset paired Wilcoxon tests with Holm correction, an across-datasets Friedman/Iman–Davenport omnibus with Nemenyi and Holm-corrected pairwise tests, and a variance decomposition. Within a single dataset, twenty seeds distinguish most method pairs (74–98%), so the protocol is not underpowered. Across the nine datasets where all 17 methods complete, the omnibus rejects global equality (Kendall’s

Item Type:	Article
Uncontrolled Keywords:	hypergraph neural networks, benchmark evaluation, statistical significance testing, reproducibility, Friedman test, critical-difference diagram, variance decomposition, node classification, leaderboard
Subjects:	Q Science > QA Mathematics > QA71-90 Instruments and machines
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Suzilawati Abu Samah
Date Deposited:	03 Sep 2026 06:10
Last Modified:	03 Sep 2026 06:10
URII:	http://shdl.mmu.edu.my/id/eprint/16613

Downloads

Edit (login required)