Chun Jiang Zhu
ABSTRACT
Hypergraphs can capture higher-order relations between subsets of objects instead of only pairwise relations as in graphs. Hypergraph clustering is an important task in information retrieval and machine learning. We study the problem of distributed hypergraph clustering in the message passing communication model using small communication cost. We propose an algorithm framework for distributed hypergraph clustering based on spectral hypergraph sparsification. For an n-vertex hypergraph G with hyperedges of maximum size r distributed at s sites arbitrarily and a parameter ε∈ (0,1), our algorithm can produce a vertex set with conductance O(√1+ε/1-ε √φG), where φG is the conductance of G, using communication cost ~O(nr2s/εO(1)) (~O hides a polylogarithmic factor). The theoretical results are complemented with extensive experiments to demonstrate the efficiency and effectiveness of the proposed algorithm under different real-world datasets. Our source code is publicly available at github.com/chunjiangzhu/dhgc.
Supplemental Material
- I. Abraham, D. Durfee, I. Koutis, S. Krinninger, and R. Peng. 2016. On fully dynamic graph sparsifiers. In Proceedings of IEEE Symposium on Foundations of Computer Science. 335--344.Google Scholar
- S. Agarwal, J. Lim, L. Zelnik-Manor, P. Perona, D. Kriegman, and S. Belongie. 2005. Beyond pairwise clustering. In Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition. 838--845.Google Scholar
- M.-F. Balcan, S. Ehrlich, and Y. Liang. 2013. Distributed k-means and k-median clustering on general communication topologies. In Proceedings of Neural Information Processing Systems Conference. 1995--2003.Google Scholar
- M.-F. Balcan, V. Kanchanapally, Y. Liang, and D. P. Woodruff. 2014. Improved distributed principal component analysis. In Proceedings of Neural Information Processing Systems Conference. 3113--3121.Google Scholar
- N. Bansal, O. Svensson, and L. Trevisan. 2019. New notions and constructions of sparsification for graphs and hypergraphs. In Proceedings of IEEE Symposium on Foundations of Computer Science. 910--928.Google Scholar
- A. R. Benson, D. F. Gleich, and J. Leskovec. 2016. Higher-order organization of complex networks. Science, Vol. 353, 6295 (2016), 163--166.Google Scholar
- S. R. Bulò and M. Pelillo. 2013. A game-theoretic approach to hypergraph clustering. IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 35, 6 (2013), 1312--1327.Google ScholarDigital Library
- T.-H. H. Chan, A. Louis, Z. G. Tang, and C. Zhang. 2018. Spectral properties of hypergraph Laplacian and approximation algorithms. J. ACM, Vol. 65, 3 (2018), 1--48.Google ScholarDigital Library
- J. Chen, H. Sun, D.P. Woodruff, and Q. Zhang. 2016. Communication-optimal distributed clustering. In Proceedings of Neural Information Processing Systems Conference. 3720--3728.Google Scholar
- Y. Chen, S. Khanna, and A. Nagda. 2020. Near-linear Size Hypergraph Cut Sparsifiers. In Proceedings of IEEE Symposium on Foundations of Computer Science. to appear.Google Scholar
- I. Chien, C.-Y. Lin, and I.-H. Wang. 2018. Community detection in hypergraphs: Optimal statistical limit and efficient algorithms. In Proceedings of International Conference on Artificial Intelligence and Statistics. 871--879.Google Scholar
- D. Ghoshdastidar and A. Dukkipati. 2014. Consistency of spectral partitioning of uniform hyper- graphs under planted partition model. In Proceedings of Neural Information Processing Systems Conference. 397--405.Google Scholar
- M. Hein, S. Setzer, L. Jost, and S. S. Rangapuram. 2013. The total variation on hypergraphs- learning on hypergraphs revisited. In Proceedings of Neural Information Processing Systems Conference. 2427--2435.Google Scholar
- M. Ikeda, A. Miyauchi, Y. Takai, and Y. Yoshida. 2018. Finding Cheeger cuts in hypergraphs via heat equation. arXiv preprint arXiv:1809.04396 (2018).Google Scholar
- M. Kapralov, R. Krauthgamer, J. Tardos, and Y. Yoshida. 2021. Towards Tight Bounds for Spectral Sparsification of Hypergraphs. In Proceedings of ACM Symposium on Theory of Computing .Google Scholar
- S. Kim, S. Nowozin, P. Kohli, and C. D. Yoo. 2011. Higher-order correlation clustering for image segmentation. In Proceedings of Neural Information Processing Systems Conference. 1530--1538.Google Scholar
- Y.T. Lee and H. Sun. 2017. An SDP-based algorithm for linear-sized spectral sparsification. In Proceedings of ACM Symposium on Theory of Computing. 678--687.Google Scholar
- M. Leordeanu and C. Sminchisescu. 2012. Efficient hypergraph clustering. In Proceedings of International Conference on Artificial Intelligence and Statistics. 676--684.Google Scholar
- P. Li, H. Dau, G. Puleo, and O. Milenkovic. 2017. Motif clustering and overlapping clustering for social network analysis. In Proceedings of IEEE International Conference on Computer Communications. 109--117.Google Scholar
- P. Li and O. Milenkovic. 2017. Inhomogeneous hypergraph clustering with applications. In Proceedings of Neural Information Processing Systems Conference. 2305--2315.Google Scholar
- T. Soma and Y. Yoshida. 2019. Spectral sparsification of hypergraphs. In Proceedings of ACM-SIAM Symposium on Discrete Algorithms. 2570--2581.Google Scholar
- D.A. Spielman and Shang-Hua Teng. 2011. Spectral sparsification of graphs. SIAM J. Comput., Vol. 40, 4 (2011), 981--1025.Google ScholarDigital Library
- H. Sun and L. Zanetti. 2019. Distributed graph clustering and sparsification. ACM Transactions on Parallel Computing, Vol. 6, 3 (2019), 17.Google ScholarDigital Library
- Y. Takai, A. Miyauchi, M. Ikeda, and Y. Yoshida. 2020. Hypergraph clustering based on PageRank. In Proceedings of ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 1970--1978.Google Scholar
- D. P. Woodruff and Q. Zhang. 2017. When distributed computation is communication expensive. Distributed Computing, Vol. 30, 5 (2017), 309--323.Google ScholarDigital Library
- H. Yin, A. R. Benson, J. Leskovec, and D. F. Gleich. 2017. Local higher-order graph clustering. In Proceedings of ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 555--564.Google Scholar
- Y. Yoshida. 2019. Cheeger inequalities for submodular transformations. In Proceedings of ACM-SIAM Symposium on Discrete Algorithms. 2582--2601.Google ScholarCross Ref
- C. Zhang, S. Hu, Z. G. Tang, and T-H. H. Chan. 2020. Re-revisiting learning on hypergraphs: Confidence interval, subgradient method, and extension to multiclass. IEEE Transactions on Knowledge and Data Engineering, Vol. 32, 3 (2020), 506--518.Google ScholarCross Ref
- D. Zhou, J. Huang, and B. Scholkopf. 2007. Learning with hypergraphs: Clustering, classification, and embedding. In Proceedings of Neural Information Processing Systems Conference. 1601--1608.Google Scholar
- C.J. Zhu, S. Storandt, K.-Y. Lam, S. Han, and J. Bi. 2019 a. Improved dynamic graph learning through fault-tolerant sparsification. In Proceedings of International Conference on Machine Learning. 7624--7633.Google Scholar
- C.J. Zhu, T. Zhu, K.-Y. Lam, S. Han, and J. Bi. 2019 b. Communication-optimal distributed dynamic graph clustering. In Proceedings of AAAI Conference on Artificial Intelligence. 5957--5964.Google Scholar
- J.Y. Zien, M.D.F. Schlag, and P.K. Chan. 1999. Multilevel spectral hypergraph partitioning with arbitrary vertex sizes. IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 18, 9 (1999), 1389--1399.Google Scholar
Index Terms
- Communication Efficient Distributed Hypergraph Clustering
Recommendations
A reductive approach to hypergraph clustering: An application to image segmentation
In the last few years, hypergraph-based methods have gained considerable attention in the resolution of real-world clustering problems, since such a mode of representation can handle higher-order relationships between elements compared to the standard ...
Iteratively Reweighted Hypergraph Subspace Clustering
CSAI '23: Proceedings of the 2023 7th International Conference on Computer Science and Artificial IntelligenceWith the development and application of hypergraphs, many hypergraph-based subspace clustering methods have been proposed by designing unique hypergraph construction based on the specific subspace representation model. However, there is no unified method ...
The Abstract Laplacian Tensor of a Hypergraph with Applications in Clustering
AbstractThe normalized abstract Laplacian tensor of a weighted hypergraph is investigated. The connectivity of the hypergraph is associated with the geometric multiplicity of the smallest eigenvalue of the abstract Laplacian tensor. There is an inequality ...
Comments