ReCycle Net: Cycle-Aware, Feature-Free GNN For Community Detection

Caleb Fernandes, Behnaz Moradi Jamei
Proceedings of GRaM: the Second Edition of the Workshop on Geometry-grounded Representation Learning and Generative Modeling, PMLR 326:184-200, 2026.

Abstract

Community detection is a fundamental problem in network science, yet classical methods suffer from resolution limits and limited adaptability, while many Graph Neural Networks (GNNs) do not explicitly model higher-order cyclic structures that underlie real-world communities. We propose ReCycle Net (RCN), a feature-free, cycle-aware GNN that integrates Renewal Non-Backtracking Random Walk (RNBRW) reinforcement into a GAT-style backbone and is trained with a multi-term unsupervised objective combining modularity, Laplacian smoothness, contrastive consistency, and orthogonality regularization. RCN targets graphs where cyclic closure is structurally informative for community formation and learns embeddings that support unsupervised community recovery. Across standard benchmarks, RCN is competitive with strong baselines (e.g., PolBooks: NMI 0.60, ARI 0.67; Facebook: silhouette 0.85), with its clearest gains appearing on overlapping protein complexes under overlap-aware evaluation (Complex Portal: ONMI 0.344 at r = 2 vs. 0.243, 0.232, and 0.140 for generative overlapping-based, strong attention-based, and modularity-based baselines). On graphs with weaker cyclic closure (e.g., Cora), gains are smaller and RCN remains comparable to standard baselines. Overall, these results suggest that explicitly incorporating cycle-derived structure into GNN learning can be beneficial in cycle-rich regimes while remaining robust outside this setting.

Cite this Paper

BibTeX
@InProceedings{pmlr-v326-fernandes26a, title = {ReCycle Net: Cycle-Aware, Feature-Free GNN For Community Detection}, author = {Fernandes, Caleb and Moradi Jamei, Behnaz}, booktitle = {Proceedings of GRaM: the Second Edition of the Workshop on Geometry-grounded Representation Learning and Generative Modeling}, pages = {184--200}, year = {2026}, editor = {Pouplin, Alison and Vadgama, Sharvaree and Bekkers, Erik and Kaba, Sékou-Oumar and Lawrence, Hannah and Lecha, Manuel and Baker, Elizabeth and Suk, Julian and Walters, Robin and Tomczak, Jakub and Jegelka, Stefanie}, volume = {326}, series = {Proceedings of Machine Learning Research}, month = {26 Apr}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v326/main/assets/fernandes26a/fernandes26a.pdf}, url = {https://proceedings.mlr.press/v326/fernandes26a.html}, abstract = {Community detection is a fundamental problem in network science, yet classical methods suffer from resolution limits and limited adaptability, while many Graph Neural Networks (GNNs) do not explicitly model higher-order cyclic structures that underlie real-world communities. We propose ReCycle Net (RCN), a feature-free, cycle-aware GNN that integrates Renewal Non-Backtracking Random Walk (RNBRW) reinforcement into a GAT-style backbone and is trained with a multi-term unsupervised objective combining modularity, Laplacian smoothness, contrastive consistency, and orthogonality regularization. RCN targets graphs where cyclic closure is structurally informative for community formation and learns embeddings that support unsupervised community recovery. Across standard benchmarks, RCN is competitive with strong baselines (e.g., PolBooks: NMI 0.60, ARI 0.67; Facebook: silhouette 0.85), with its clearest gains appearing on overlapping protein complexes under overlap-aware evaluation (Complex Portal: ONMI 0.344 at r = 2 vs. 0.243, 0.232, and 0.140 for generative overlapping-based, strong attention-based, and modularity-based baselines). On graphs with weaker cyclic closure (e.g., Cora), gains are smaller and RCN remains comparable to standard baselines. Overall, these results suggest that explicitly incorporating cycle-derived structure into GNN learning can be beneficial in cycle-rich regimes while remaining robust outside this setting.} }
Endnote
%0 Conference Paper %T ReCycle Net: Cycle-Aware, Feature-Free GNN For Community Detection %A Caleb Fernandes %A Behnaz Moradi Jamei %B Proceedings of GRaM: the Second Edition of the Workshop on Geometry-grounded Representation Learning and Generative Modeling %C Proceedings of Machine Learning Research %D 2026 %E Alison Pouplin %E Sharvaree Vadgama %E Erik Bekkers %E Sékou-Oumar Kaba %E Hannah Lawrence %E Manuel Lecha %E Elizabeth Baker %E Julian Suk %E Robin Walters %E Jakub Tomczak %E Stefanie Jegelka %F pmlr-v326-fernandes26a %I PMLR %P 184--200 %U https://proceedings.mlr.press/v326/fernandes26a.html %V 326 %X Community detection is a fundamental problem in network science, yet classical methods suffer from resolution limits and limited adaptability, while many Graph Neural Networks (GNNs) do not explicitly model higher-order cyclic structures that underlie real-world communities. We propose ReCycle Net (RCN), a feature-free, cycle-aware GNN that integrates Renewal Non-Backtracking Random Walk (RNBRW) reinforcement into a GAT-style backbone and is trained with a multi-term unsupervised objective combining modularity, Laplacian smoothness, contrastive consistency, and orthogonality regularization. RCN targets graphs where cyclic closure is structurally informative for community formation and learns embeddings that support unsupervised community recovery. Across standard benchmarks, RCN is competitive with strong baselines (e.g., PolBooks: NMI 0.60, ARI 0.67; Facebook: silhouette 0.85), with its clearest gains appearing on overlapping protein complexes under overlap-aware evaluation (Complex Portal: ONMI 0.344 at r = 2 vs. 0.243, 0.232, and 0.140 for generative overlapping-based, strong attention-based, and modularity-based baselines). On graphs with weaker cyclic closure (e.g., Cora), gains are smaller and RCN remains comparable to standard baselines. Overall, these results suggest that explicitly incorporating cycle-derived structure into GNN learning can be beneficial in cycle-rich regimes while remaining robust outside this setting.
APA
Fernandes, C. & Moradi Jamei, B.. (2026). ReCycle Net: Cycle-Aware, Feature-Free GNN For Community Detection. Proceedings of GRaM: the Second Edition of the Workshop on Geometry-grounded Representation Learning and Generative Modeling, in Proceedings of Machine Learning Research 326:184-200 Available from https://proceedings.mlr.press/v326/fernandes26a.html.

Related Material