Learning Directed Knowledge Using Higher-Ordered Neural Networks: Building a Predictive Framework
Source abstract
Abstract
Most graph learning methods remain limited to undirected, pairwise interactions, restricting their ability to capture the multi-entity and directional relationships common in real-world systems. We propose the Directed Higher-Ordered Neural Network (HONN) framework that introduces directionality into hypergraph learning through flexible spectral Laplacian formulations. Unlike fixed-Laplacian methods such as the Generalized Directed Hypergraph Neural Network (GeDi-HNN), a tunable q-parameter in our framework balances local identity preservation with global diffusion, enabling robust and generalizable feature propagation. Experiments on five benchmark datasets show that HONN consistently matches or outperforms state-of-the-art baselines, achieving 84% on NTU-2012, 87.4% on WebKB Texas, and 86.2% on Cornell, while maintaining computational efficiency. Ablation studies confirm the crucial role of Laplacian selection, activation functions, and q-tuning in shaping model performance. By unifying directionality and higher-order reasoning, HONN provides a scalable foundation for predictive modeling in domains such as knowledge graphs, spatio-temporal networks, and recommendation systems.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Advanced Graph Neural Networks · Domain Adaptation and Few-Shot Learning · Topic Modeling
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Systems and Data Engineering Team (IDS), National School of Applied Sciences, Abdelmalek Essaadi University, Tangier 93000, Morocco
LISV Laboratory, University of Paris-Saclay, 10–12 Avenue of Europe, 78140 Velizy, France
École Centrale d’Électronique (ECE Engineering School), 10 Rue Sextius Michel, 75015 Paris, France
Faculty of Electrical Engineering, Electronics and Automation, University of Ruse “Angel Kanchev”, 8 Studentska Street, 7017 Ruse, Bulgaria
Funding
European Union—NextGenerationEU
References (41)
Show source-provided references
1. Machine learning on graphs: A model and comprehensive taxonomy
2. A comprehensive survey on graph neural networks10.1109/TNNLS.2020.2978386 ↗
3. The graph neural network model10.1109/TNN.2008.2005605 ↗
4. Kipf, T.N., and Welling, M. (2016). Semi-supervised classification with graph convolutional networks. arXiv.
5. Shchur, O., Mumme, M., Bojchevski, A., and Günnemann, S. (2018). Pitfalls of graph neural network evaluation. arXiv.
6. A directed hypergraph model for random time dependent shortest paths
7. Linear connectivity problems in directed hypergraphs10.1016/j.tcs.2009.02.038 ↗
8. Tran, H.Q., and Lin, Y. (2023, January 17–24). Directed hypergraph neural networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Vancouver, BC, Canada.
9. Learning with hypergraphs: Clustering, classification, and embedding
10. Neuro-symbolic artificial intelligence: A survey10.1007/s00521-024-09960-z ↗
11. Bhuyan, B.P., Ramdane-Cherif, A., Singh, T.P., and Tomar, R. (2025). Studies in Computational Intelligence. Neuro-Symbolic Artificial Intelligence: Bridging Logic and Learning, Springer Nature.10.1007/978-981-97-8171-3 ↗
12. MagNet: A neural network for directed graphs
13. SigMaNet: One Laplacian to rule them all
14. Hypergraph convolution and hypergraph attention
15. Hypergraph neural networks
16. Veličković, P., Cucurull, G., Casanova, A., Romero, A., Lio, P., and Bengio, Y. (May, January 30). Graph attention networks. Proceedings of the International Conference on Learning Representations (ICLR), Vancouver, BC, Canada.
17. Directed hypergraph neural network10.5373/JARDCS/V12SP4/20201622 ↗
18. Fiorini, S., Coniglio, S., Ciavotta, M., and Del Bue, A. (2024). Let There be Direction in Hypergraph Neural Networks. Trans. Mach. Learn. Res., Available online: https://openreview.net/forum?id=h48Ri6pmvi.
19. Unequal error protection with the H.264 flexible macroblock ordering
20. A General Method for Generating Discrete Orthogonal Matrices10.1109/ACCESS.2021.3107579 ↗
21. Lin, X., Zhang, W., Shi, F., Zhou, C., Zou, L., Zhao, X., Yin, D., Pan, S., and Cao, Y. (2024, January 21–27). Graph Neural Stochastic Diffusion for Estimating Uncertainty in Node Classification. Proceedings of the 41st International Conference on Machine Learning (ICML), Vienna, Austria.
22. Yang, M., and Xu, X.-J. (2025). Recent Advances in Hypergraph Neural Networks. arXiv.10.1007/s40305-025-00630-y ↗
23. Ko, T., Choi, Y., and Kim, C.-K. (August, January 31). Universal Graph Contrastive Learning with a Novel Laplacian Perturbation. Proceedings of the Thirty-Ninth Conference on Uncertainty in Artificial Intelligence (UAI), Pittsburgh, PA, USA.
24. Networks beyond pairwise interactions: Structure and dynamics10.1016/j.physrep.2020.05.004 ↗
25. Berge, C. (1984). Hypergraphs: Combinatorics of finite sets. North-Holland Mathematical Library, Elsevier.
26. Mathematical Engineering
27. Hypergraph convolution and hypergraph attention10.1016/j.patcog.2020.107637 ↗
28. Directed hypergraphs: Introduction and fundamental algorithms—A survey10.1016/j.tcs.2016.03.016 ↗
29. On the orientation of graphs and hypergraphs10.1016/S0166-218X(02)00462-6 ↗
30. Transferability of spectral graph convolutional neural networks
31. HpLapGCN: Hypergraph p-Laplacian graph convolutional networks10.1016/j.neucom.2019.06.068 ↗
32. HGNN+: General hypergraph neural networks10.1109/TPAMI.2022.3182052 ↗
33. Handscomb, D.C., and Mason, J.C. (2002). Chebyshev Polynomials, Chapman and Hall/CRC.
34. Singh, R., and Chen, Y. (2022). Signed graph neural networks: A frequency perspective. arXiv.
35. Chung, F.R.K. (1997). Spectral Graph Theory, American Mathematical Society.
36. Zhang, J., Hui, B., Harn, P.-W., Sun, M.-T., and Ku, W.-S. (2021). smgc: A complex-valued graph convolutional network via magnetic Laplacian for directed graphs. arXiv.
37. Smith, L.N. (2019, January 16–17). Super-convergence: Very fast training of neural networks using large learning rates. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, Long Beach, CA, USA.10.1117/12.2520589 ↗
38. Goyal, P., Ferrara, E., and Le, H. (2017, January 6–11). Accurate learning rate scheduling for deep networks. Proceedings of the IEEE International Conference on Machine Learning (ICML), Sydney, Australia.
39. Yan, Y., Hashemi, M., Swersky, K., and Yang, Y. (2010). Two sides of the same coin: Heterophily and oversmoothing in graph convolutional neural networks. arXiv.
40. Activation functions and their characteristics in deep neural networks
41. Hu, W., Fey, M., Zitnik, M., Dong, Y., Ren, H., Liu, B., Catasta, M., and Leskovec, J. (2020). Open Graph Benchmark: Datasets for Machine Learning on Graphs. arXiv.