{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,4]],"date-time":"2025-11-04T10:35:06Z","timestamp":1762252506402,"version":"build-2065373602"},"reference-count":50,"publisher":"Springer Science and Business Media LLC","issue":"8","license":[{"start":{"date-parts":[[2025,8,27]],"date-time":"2025-08-27T00:00:00Z","timestamp":1756252800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2025,8,27]],"date-time":"2025-08-27T00:00:00Z","timestamp":1756252800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100005366","name":"University of Oslo","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100005366","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Int J Data Sci Anal"],"published-print":{"date-parts":[[2025,12]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Graph neural networks (GNNs) have achieved significant success in various applications. Most GNNs learn the node features with information aggregation of its neighbors and feature transformation in each layer. However, the node features become indistinguishable after many layers, leading to performance deterioration: a significant limitation known as over-smoothing. Past work adopted various techniques for addressing this issue, such as normalization and skip connection of layer-wise output. After the study, we found that the information aggregations in existing work are all contracted aggregations, with the intrinsic property that features will inevitably converge to the same single point after many layers. To this end, we propose the aggregation over compacted manifold (ACM) method that replaces the existing information aggregation with aggregation over compact manifolds, a special type of manifold, which avoids contracted aggregations. In this work, we theoretically analyze contracted aggregation and its properties. We also provide an extensive empirical evaluation that shows ACM can effectively alleviate over-smoothing and outperforms the state of the art. The code can be found in\n                    <jats:ext-link xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" xlink:href=\"https:\/\/github.com\/DongzhuoranZhou\/ACM.git\" ext-link-type=\"uri\">https:\/\/github.com\/DongzhuoranZhou\/ACM.git<\/jats:ext-link>\n                    .\n                  <\/jats:p>","DOI":"10.1007\/s41060-025-00875-z","type":"journal-article","created":{"date-parts":[[2025,8,27]],"date-time":"2025-08-27T11:52:09Z","timestamp":1756295529000},"page":"7055-7069","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Alleviating over-smoothing via aggregation over compact manifolds (extended version)"],"prefix":"10.1007","volume":"20","author":[{"given":"Dongzhuoran","family":"Zhou","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hui","family":"Yang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Bo","family":"Xiong","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yue","family":"Ma","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Evgeny","family":"Kharlamov","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2025,8,27]]},"reference":[{"key":"875_CR1","doi-asserted-by":"publisher","first-page":"57","DOI":"10.1016\/j.aiopen.2021.01.001","volume":"1","author":"J Zhou","year":"2020","unstructured":"Zhou, J., Cui, G., et al.: Graph neural networks: a review of methods and applications. AI Open 1, 57\u201381 (2020)","journal-title":"AI Open"},{"unstructured":"Xu, K., Hu, W., et al.: How powerful are graph neural networks? In: ICLR (2019)","key":"875_CR2"},{"unstructured":"Hamilton, W.L., et al: Inductive representation learning on large graphs. In: NIPS, pp. 1024\u20131034 (2017)","key":"875_CR3"},{"doi-asserted-by":"crossref","unstructured":"Gao, H., Wang, Z., Ji, S.: Large-scale learnable graph convolutional networks. In: Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 1416\u20131424 (2018)","key":"875_CR4","DOI":"10.1145\/3219819.3219947"},{"doi-asserted-by":"crossref","unstructured":"Hamilton, W.L.: Graph Representation Learning. In: Synthesis Lectures on Artificial Intelligence and Machine Learning (2020)","key":"875_CR5","DOI":"10.1007\/978-3-031-01588-5"},{"doi-asserted-by":"crossref","unstructured":"Li, Q., Han, Z., Wu, X.-M.: Deeper insights into graph convolutional networks for semi-supervised learning. In: AAAI, pp. 3538\u20133545 (2018)","key":"875_CR6","DOI":"10.1609\/aaai.v32i1.11604"},{"unstructured":"Ioffe, S., Szegedy, C.: Batch normalization: Accelerating deep network training by reducing internal covariate shift. In: ICML, pp. 448\u2013456 (2015)","key":"875_CR7"},{"unstructured":"Zhao, L., Akoglu, L.: Pairnorm: Tackling oversmoothing in GNNs. In: ICLR (2020)","key":"875_CR8"},{"unstructured":"Zhou, K., Et al.: Towards deeper graph neural networks with differentiable group normalization. In: NeurIPS (2020)","key":"875_CR9"},{"unstructured":"Rong, Y., Huang, W., Xu, T., Huang, J.: Dropedge: Towards deep graph convolutional networks on node classification. In: ICLR (2020)","key":"875_CR10"},{"unstructured":"Chen, D., Lin, Y., Et al.: Measuring and relieving the over-smoothing problem for graph neural networks from the topological view. CoRR arXiv:1909.03211 (2019)","key":"875_CR11"},{"unstructured":"Hou, Y., Zhang, J., Et al.: Measuring and improving the use of graph information in graph neural networks. In: ICLR (2020)","key":"875_CR12"},{"doi-asserted-by":"crossref","unstructured":"Jin, W., Liu, X., Ma, Y., Aggarwal, C., Tang, J.: Feature overcorrelation in deep graph neural networks: A new perspective. arXiv preprint arXiv:2206.07743 (2022)","key":"875_CR13","DOI":"10.1145\/3534678.3539445"},{"unstructured":"Xu, K., Li, C., Tian, Y., Et al.: Representation learning on graphs with jumping knowledge networks. In: ICML, pp. 5449\u20135458. PMLR (2018)","key":"875_CR14"},{"doi-asserted-by":"crossref","unstructured":"Li, G., Muller, M., Thabet, A., Ghanem, B.: Deepgcns: Can gcns go as deep as CNNs? In: Proceedings of the IEEE\/CVF International Conference on Computer Vision, pp. 9267\u20139276 (2019)","key":"875_CR15","DOI":"10.1109\/ICCV.2019.00936"},{"unstructured":"Huang, W., Et al.: Tackling over-smoothing for general graph convolutional networks. CoRR (2020)","key":"875_CR16"},{"unstructured":"Oono, K., Suzuki, T.: Graph neural networks exponentially lose expressive power for node classification. In: ICLR (2020)","key":"875_CR17"},{"unstructured":"Cai, C., Wang, Y.: A note on over-smoothing for graph neural networks. CoRR abs\/2006.13318 (2020)","key":"875_CR18"},{"doi-asserted-by":"crossref","unstructured":"He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770\u2013778 (2016)","key":"875_CR19","DOI":"10.1109\/CVPR.2016.90"},{"unstructured":"Chen, M., Wei, Z., Huang, Z., Ding, B., Li, Y.: Simple and deep graph convolutional networks. In: ICML, pp. 1725\u20131735. PMLR (2020)","key":"875_CR20"},{"issue":"1","key":"875_CR21","first-page":"1929","volume":"15","author":"N Srivastava","year":"2014","unstructured":"Srivastava, N., et al.: Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15(1), 1929\u20131958 (2014)","journal-title":"J. Mach. Learn. Res."},{"unstructured":"Klicpera, J., Et al.: Predict then propagate: Graph neural networks meet personalized pagerank. In: ICLR (2019)","key":"875_CR22"},{"unstructured":"Chien, E., Peng, J., Li, P., Milenkovic, O.: Adaptive universal generalized pagerank graph neural network. In: ICLR (2021)","key":"875_CR23"},{"doi-asserted-by":"crossref","unstructured":"Liu, M., Gao, H., Ji, S.: Towards deeper graph neural networks. In: Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 338\u2013348 (2020)","key":"875_CR24","DOI":"10.1145\/3394486.3403076"},{"unstructured":"Balazevic, I., Allen, C., Hospedales, T.M.: Multi-relational poincar\u00e9 graph embeddings. In: NeurIPS, pp. 4465\u20134475 (2019)","key":"875_CR25"},{"unstructured":"G\u00fcl\u00e7ehre, \u00c7., et al.: Hyperbolic attention networks. In: ICLR (2019)","key":"875_CR26"},{"doi-asserted-by":"crossref","unstructured":"Khrulkov, V., Mirvakhabova, L., Ustinova, E., Oseledets, I., Lempitsky, V.: Hyperbolic image embeddings. In: Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, pp. 6418\u20136428 (2020)","key":"875_CR27","DOI":"10.1109\/CVPR42600.2020.00645"},{"unstructured":"Chami, I., et al: Hyperbolic graph convolutional neural networks. In: NeurIPS, pp. 4869\u20134880 (2019)","key":"875_CR28"},{"key":"875_CR29","doi-asserted-by":"publisher","first-page":"4633","DOI":"10.1609\/aaai.v37i4.25586","volume":"37","author":"L Sun","year":"2023","unstructured":"Sun, L., Ye, J., Peng, H., Wang, F., Yu, P.S.: Self-supervised continual graph learning in adaptive riemannian spaces. Proceedings of the AAAI Conference on Artificial Intelligence 37, 4633\u20134642 (2023)","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"key":"875_CR30","doi-asserted-by":"publisher","first-page":"9044","DOI":"10.1609\/aaai.v38i8.28754","volume":"38","author":"L Sun","year":"2024","unstructured":"Sun, L., Huang, Z., Wang, Z., Wang, F., Peng, H., Yu, P.S.: Motif-aware riemannian graph neural network with generative-contrastive learning. Proceedings of the AAAI Conference on Artificial Intelligence 38, 9044\u20139052 (2024)","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"unstructured":"Bachmann, G., B\u00e9cigneul, G., Ganea, O.: Constant curvature graph convolutional networks. In: ICML, pp. 486\u2013496. PMLR (2020)","key":"875_CR31"},{"unstructured":"Ganea, O., B\u00e9cigneul, G., Hofmann, T.: Hyperbolic neural networks. In: NeurIPS, pp. 5350\u20135360 (2018)","key":"875_CR32"},{"doi-asserted-by":"crossref","unstructured":"Unnikrishnan, N.K., Gould, J., Parhi, K.K.: Scv-gnn: Sparse compressed vector-based graph neural network aggregation. IEEE Trans. Comput. Aided Des. Integr. Circuits Syst. 42(12), 4803\u20134816 (2023)","key":"875_CR33","DOI":"10.1109\/TCAD.2023.3291672"},{"unstructured":"Oc\u00e1riz\u00a0Borde, H.S., Kratsios, A.: Neural snowflakes: Universal latent graph inference via trainable latent geometries. In: The Twelfth International Conference on Learning Representations (2023)","key":"875_CR34"},{"unstructured":"Donnat, C., Jeong, S.W.: Studying the effect of gnn spatial convolutions on the embedding space\u2019s geometry. In: Uncertainty in Artificial Intelligence, pp. 539\u2013548. PMLR (2023)","key":"875_CR35"},{"doi-asserted-by":"crossref","unstructured":"Wang, B., Jiang, B.: Generalizing aggregation functions in gnns: High-capacity gnns via nonlinear neighborhood aggregators. arXiv preprint arXiv:2202.09145 (2022)","key":"875_CR36","DOI":"10.1109\/TPAMI.2023.3290649"},{"key":"875_CR37","doi-asserted-by":"publisher","first-page":"850","DOI":"10.1109\/TSIPN.2023.3341408","volume":"9","author":"A Parada-Mayorga","year":"2023","unstructured":"Parada-Mayorga, A., Wang, Z., Gama, F., Ribeiro, A.: Stability of aggregation graph neural networks. IEEE Transactions on Signal and Information Processing over Networks 9, 850\u2013864 (2023)","journal-title":"IEEE Transactions on Signal and Information Processing over Networks"},{"unstructured":"Mendelson, B.: Introduction to Topology (1990)","key":"875_CR38"},{"key":"875_CR39","volume-title":"Introduction to Smooth Manifolds","author":"JM Lee","year":"2013","unstructured":"Lee, J.M.: Introduction to Smooth Manifolds. Graduate Texts in Mathematics. Springer, Berlin (2013)"},{"unstructured":"Kipf, T.N., Welling, M.: Semi-supervised classification with graph convolutional networks. CoRR arXiv:1609.02907 (2016)","key":"875_CR40"},{"unstructured":"Velickovic, P., Cucurull, G., Casanova, A., Romero, A., Li\u00f2, P., Bengio, Y.: Graph attention networks. CoRR arXiv:1710.10903 (2017)","key":"875_CR41"},{"unstructured":"Wu, F., Souza, A., Zhang, T., Fifty, C., Yu, T., Weinberger, K.: Simplifying graph convolutional networks. In: International Conference on Machine Learning, pp. 6861\u20136871. PMLR (2019)","key":"875_CR42"},{"doi-asserted-by":"crossref","unstructured":"Ungar, A.A.: Barycentric Calculus in Euclidean and Hyperbolic Geometry: A Comparative Introduction (2010)","key":"875_CR43","DOI":"10.1142\/9789814304948"},{"unstructured":"Yang, Z., et al.: Revisiting semi-supervised learning with graph embeddings. In: ICML. JMLR Workshop and Conference Proceedings 48, 40\u201348 (2016)","key":"875_CR44"},{"unstructured":"Shchur, O., Mumme, M., Bojchevski, A., G\u00fcnnemann, S.: Pitfalls of graph neural network evaluation. CoRR arXiv:1811.05868 (2018)","key":"875_CR45"},{"unstructured":"Pei, H., et al.: Geom-gcn: Geometric graph convolutional networks. In: ICLR (2020)","key":"875_CR46"},{"doi-asserted-by":"crossref","unstructured":"Rashid, A.M., Karypis, G., et al.: Learning preferences of new users in recommender systems: an information theoretic approach. In: SIGKDD Explor., pp. 90\u2013100 (2008)","key":"875_CR47","DOI":"10.1145\/1540276.1540302"},{"unstructured":"Klicpera, J., et al: Predict then propagate: Graph neural networks meet personalized pagerank. In: ICLR (2019)","key":"875_CR48"},{"unstructured":"Kingma, D.P., Ba, J.: Adam: A method for stochastic optimization. In: ICLR (2015)","key":"875_CR49"},{"unstructured":"Glorot, X., Bengio, Y.: Understanding the difficulty of training deep feedforward neural networks. In: AISTATS, pp. 249\u2013256 (2010)","key":"875_CR50"}],"container-title":["International Journal of Data Science and Analytics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s41060-025-00875-z.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s41060-025-00875-z\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s41060-025-00875-z.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,4]],"date-time":"2025-11-04T10:31:01Z","timestamp":1762252261000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s41060-025-00875-z"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,8,27]]},"references-count":50,"journal-issue":{"issue":"8","published-print":{"date-parts":[[2025,12]]}},"alternative-id":["875"],"URL":"https:\/\/doi.org\/10.1007\/s41060-025-00875-z","relation":{},"ISSN":["2364-415X","2364-4168"],"issn-type":[{"type":"print","value":"2364-415X"},{"type":"electronic","value":"2364-4168"}],"subject":[],"published":{"date-parts":[[2025,8,27]]},"assertion":[{"value":"16 December 2024","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"9 July 2025","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"27 August 2025","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no Conflict of interest.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflict of interest"}}]}}