Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
97 tokens/sec
GPT-4o
53 tokens/sec
Gemini 2.5 Pro Pro
44 tokens/sec
o3 Pro
5 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

GraphRARE: Reinforcement Learning Enhanced Graph Neural Network with Relative Entropy (2312.09708v2)

Published 15 Dec 2023 in cs.LG and cs.AI

Abstract: Graph neural networks (GNNs) have shown advantages in graph-based analysis tasks. However, most existing methods have the homogeneity assumption and show poor performance on heterophilic graphs, where the linked nodes have dissimilar features and different class labels, and the semantically related nodes might be multi-hop away. To address this limitation, this paper presents GraphRARE, a general framework built upon node relative entropy and deep reinforcement learning, to strengthen the expressive capability of GNNs. An innovative node relative entropy, which considers node features and structural similarity, is used to measure mutual information between node pairs. In addition, to avoid the sub-optimal solutions caused by mixing useful information and noises of remote nodes, a deep reinforcement learning-based algorithm is developed to optimize the graph topology. This algorithm selects informative nodes and discards noisy nodes based on the defined node relative entropy. Extensive experiments are conducted on seven real-world datasets. The experimental results demonstrate the superiority of GraphRARE in node classification and its capability to optimize the original graph topology.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (60)
  1. S. Abu-El-Haija, B. Perozzi, A. Kapoor, N. Alipourfard, K. Lerman, H. Harutyunyan, G. Ver Steeg, and A. Galstyan, “Mixhop: Higher-order graph convolutional architectures via sparsified neighborhood mixing,” in ICML, 2019, pp. 21–29.
  2. G. Brockman, V. Cheung, L. Pettersson, J. Schneider, J. Schulman, J. Tang, and W. Zaremba, “Openai gym,” 2016.
  3. T. M. Cover, J. A. Thomas et al., “Entropy, relative entropy and mutual information,” Elements of information theory, vol. 2, no. 1, pp. 12–13, 1991.
  4. L. Du, X. Shi, Q. Fu, X. Ma, H. Liu, S. Han, and D. Zhang, “GBK-GNN: gated bi-kernel graph neural networks for modeling both homophily and heterophily,” in WWW, 2022, pp. 1550–1558.
  5. J. Fang, Y. Shen, Y. Wang, and L. Chen, “Optimizing DNN computation graph using graph substitutions,” Proc. VLDB Endow., vol. 13, no. 11, pp. 2734–2746, 2020.
  6. Z. Fang, L. Xu, G. Song, Q. Long, and Y. Zhang, “Polarized graph neural networks,” in WWW, 2022, pp. 1404–1413.
  7. K. Feng, C. Li, X. Zhang, and J. Zhou, “Towards open temporal graph neural networks,” arXiv preprint arXiv:2303.15015, 2023.
  8. M. Fey and J. E. Lenssen, “Fast graph representation learning with PyTorch Geometric,” in ICLR Workshop on Representation Learning on Graphs and Manifolds, 2019, pp. 1–12.
  9. A. P. García-Plaza, V. Fresno-Fernández, R. Martínez-Unanue, and A. Zubiaga, “Using fuzzy logic to leverage HTML markup for web page representation,” IEEE Trans. Fuzzy Syst., vol. 25, no. 4, pp. 919–933, 2017.
  10. Z. Gu, K. Zhang, G. Bai, L. Chen, L. Zhao, and C. Yang, “Dynamic activation of clients and parameters for federated learning over heterogeneous graphs,” in ICDE, 2023, pp. 1597–1610.
  11. W. L. Hamilton, Z. Ying, and J. Leskovec, “Inductive representation learning on large graphs,” in NeurIPS, vol. 30, 2017, pp. 1024–1034.
  12. Y. Han, G. Li, H. Yuan, and J. Sun, “An autonomous materialized view management system with deep reinforcement learning,” in ICDE, 2021, pp. 2159–2164.
  13. D. He, C. Liang, H. Liu, M. Wen, P. Jiao, and Z. Feng, “Block modeling-guided graph convolutional neural networks,” in AAAI, 2022, pp. 4022–4029.
  14. S. Horchidan, “Query optimization for inference-based graph databases,” in VLDB, ser. CEUR Workshop Proceedings, V. Efthymiou and X. Hu, Eds., vol. 3452, 2023, pp. 33–36.
  15. Z. Jia, J. Thomas, T. Warszawski, M. Gao, M. Zaharia, and A. Aiken, “Optimizing DNN computation with relaxed graph substitutions,” in Proceedings of Machine Learning and Systems, 2019.
  16. D. Jin, Z. Yu, C. Huo, R. Wang, X. Wang, D. He, and J. Han, “Universal graph convolutional networks,” in NeurIPS, vol. 34, 2021, pp. 10 654–10 664.
  17. W. Jin, T. Derr, Y. Wang, Y. Ma, Z. Liu, and J. Tang, “Node similarity preserving graph convolutional networks,” in WSDM, 2021, pp. 148–156.
  18. T. N. Kipf and M. Welling, “Semi-supervised classification with graph convolutional networks,” in ICLR, 2017, pp. 1–14.
  19. S. Kullback and R. A. Leibler, “On information and sufficiency,” The annals of mathematical statistics, vol. 22, no. 1, pp. 79–86, 1951.
  20. C. Li, Y. Tsai, and J. C. Liao, “Graph neural networks for tabular data learning,” in ICDE, 2023, pp. 3589–3592.
  21. Y. Li, H. Yuan, Z. Fu, X. Ma, M. Xu, and S. Wang, “ELASTIC: edge workload forecasting based on collaborative cloud-edge deep learning,” in WWW, Y. Ding, J. Tang, J. F. Sequeda, L. Aroyo, C. Castillo, and G. Houben, Eds., 2023, pp. 3056–3066.
  22. Y. Li, H. Xiong, L. Kong, S. Wang, Z. Sun, H. Chen, G. Chen, and D. Yin, “Ltrgcn: Large-scale graph convolutional networks-based learning to rank for web search,” in Joint European Conference on Machine Learning and Knowledge Discovery in Databases, 2023, pp. 635–651.
  23. Y. Li, H. Xiong, L. Kong, R. Zhang, F. Xu, G. Chen, and M. Li, “Mhrr: Moocs recommender service with meta hierarchical reinforced ranking,” IEEE Transactions on Services Computing, 2023.
  24. J. Lin, “Divergence measures based on the shannon entropy,” IEEE Transactions on Information theory, vol. 37, no. 1, pp. 145–151, 1991.
  25. M. Liu, Z. Wang, and S. Ji, “Non-local graph neural networks,” 2021.
  26. G. Luo, J. Li, H. Peng, C. Yang, L. Sun, P. S. Yu, and L. He, “Graph entropy guided node embedding dimension selection for graph neural networks,” IJCAI, pp. 2767–2774, 2021.
  27. X. Ma, Q. Chen, Y. Ren, G. Song, and L. Wang, “Meta-weight graph neural network: Push the limits beyond global homophily,” in WWW, 2022, pp. 1270–1280.
  28. G. Namata, B. London, L. Getoor, B. Huang, and U. Edu, “Query-driven active surveying for collective classification,” in 10th International Workshop on Mining and Learning with Graphs, vol. 8, 2012, p. 1.
  29. Y. M. Omar and P. Plapper, “A survey of information entropy metrics for complex networks,” Entropy, vol. 22, no. 12, p. 1417, 2020.
  30. A. Paszke, S. Gross, S. Chintala, G. Chanan, E. Yang, Z. DeVito, Z. Lin, A. Desmaison, L. Antiga, and A. Lerer, “Automatic differentiation in pytorch,” NeurIPS, 2017.
  31. H. Pei, B. Wei, K. C. Chang, Y. Lei, and B. Yang, “Geom-gcn: Geometric graph convolutional networks,” in ICLR, 2020, pp. 1–12.
  32. T. Peng, Y. Liang, W. Wu, J. Ren, Z. Pengrui, and Y. Pu, “Clgt: A graph transformer for student performance prediction in collaborative learning,” in AAAI, vol. 37, no. 13, 2023, pp. 15 947–15 954.
  33. A. Raffin, A. Hill, A. Gleave, A. Kanervisto, M. Ernestus, and N. Dormann, “Stable-baselines3: Reliable reinforcement learning implementations,” JMLR, vol. 22, no. 268, pp. 1–8, 2021.
  34. B. Rozemberczki, C. Allen, and R. Sarkar, “Multi-scale attributed node embedding,” Journal of Complex Networks, vol. 9, no. 2, p. cnab014, 2021.
  35. J. Schulman, F. Wolski, P. Dhariwal, A. Radford, and O. Klimov, “Proximal policy optimization algorithms,” 2017.
  36. P. Sen, G. Namata, M. Bilgic, L. Getoor, B. Galligher, and T. Eliassi-Rad, “Collective classification in network data,” AI magazine, vol. 29, no. 3, pp. 93–93, 2008.
  37. Z. Shen, C. Hu, and Z. Zhao, “Lynx: A graph query framework for multiple heterogeneous data sources,” Proc. VLDB Endow., vol. 16, no. 12, pp. 3926–3929, 2023.
  38. L. Tian and H. Wu, “MI-GCN: node mutual information-based graph convolutional network,” in WWW, 2022, pp. 996–1003.
  39. N. Vasilache, O. Zinenko, T. Theodoridis, P. Goyal, Z. DeVito, W. S. Moses, S. Verdoolaege, A. Adams, and A. Cohen, “Tensor comprehensions: Framework-agnostic high-performance machine learning abstractions,” CoRR, vol. abs/1802.04730, 2018.
  40. P. Velickovic, G. Cucurull, A. Casanova, A. Romero, P. Liò, and Y. Bengio, “Graph attention networks,” in ICLR, 2018, pp. 1–12.
  41. M. Wang, D. Zheng, Z. Ye, Q. Gan, M. Li, X. Song, J. Zhou, C. Ma, L. Yu, Y. Gai et al., “Deep graph library: A graph-centric, highly-performant package for graph neural networks,” 2019.
  42. T. Wang, D. Jin, R. Wang, D. He, and Y. Huang, “Powerful graph convolutional networks with adaptive propagation mechanism for homophily and heterophily,” in AAAI, 2022, pp. 4210–4218.
  43. C. Wu, C. Wang, J. Xu, Z. Fang, T. Gu, C. Wang, Y. Song, K. Zheng, X. Wang, and G. Zhou, “Instant representation learning for recommendation over large dynamic graphs,” in ICDE, 2023, pp. 82–95.
  44. Y. Yan, M. Hashemi, K. Swersky, Y. Yang, and D. Koutra, “Two sides of the same coin: Heterophily and oversmoothing in graph convolutional neural networks,” in ICDM, 2022, pp. 1287–1292.
  45. T. Yang, Y. Wang, Z. Yue, Y. Yang, Y. Tong, and J. Bai, “Graph pointer neural networks,” in AAAI, 2022, pp. 8832–8839.
  46. Y. Yang, Z. Guan, J. Li, W. Zhao, J. Cui, and Q. Wang, “Interpretable and efficient heterogeneous graph convolutional network,” TKDE, vol. 35, no. 2, pp. 1637–1650, 2023.
  47. H. Yuan and G. Li, “A survey of traffic prediction: from spatio-temporal data to intelligent transportation,” Data Sci. Eng., vol. 6, no. 1, pp. 63–85, 2021.
  48. H. Yuan, G. Li, and Z. Bao, “Route travel time estimation on A road network revisited: Heterogeneity, proximity, periodicity and dynamicity,” Proc. VLDB Endow., vol. 16, no. 3, pp. 393–405, 2022.
  49. H. Yuan, G. Li, Z. Bao, and L. Feng, “An effective joint prediction model for travel demands and traffic flows,” in ICDE, 2021, pp. 348–359.
  50. Q. Zhang, M. Li, and Y. Deng, “Measure the structure similarity of nodes in complex networks based on relative entropy,” Physica A: Statistical Mechanics and its Applications, vol. 491, pp. 749–763, 2018.
  51. Y. Zhang, Q. Yao, L. Yue, X. Wu, Z. Zhang, Z. Lin, and Y. Zheng, “Emerging drug interaction prediction enabled by flow-based graph neural network with biomedical network,” CoRR, vol. abs/2311.09261, 2023.
  52. Y. Zhang, Z. Zhou, Q. Yao, X. Chu, and B. Han, “Adaprop: Learning adaptive propagation for graph neural network based knowledge graph reasoning,” in SIGKDD.   ACM, 2023, pp. 3446–3457.
  53. Y. Zhang, W. Wang, H. Yin, P. Zhao, W. Chen, and L. Zhao, “Disconnected emerging knowledge graph oriented inductive link prediction,” in ICDE, 2023, pp. 381–393.
  54. H. Zhao, Q. Yao, and W. Tu, “Search to aggregate neighborhood for graph neural network,” in ICDE, 2021, pp. 552–563.
  55. J. Zhao, Z. Zhou, Z. Guan, W. Zhao, W. Ning, G. Qiu, and X. He, “Intentgc: A scalable graph convolution framework fusing heterogeneous information for recommendation,” in SIGKDD, 2019, pp. 2347–2357.
  56. S. Zheng, W. Wang, J. Qu, H. Yin, W. Chen, and L. Zhao, “MMKGR: multi-hop multi-modal knowledge graph reasoning,” in ICDE, 2023, pp. 96–109.
  57. X. Zheng, Y. Liu, S. Pan, M. Zhang, D. Jin, and P. S. Yu, “Graph neural networks for graphs with heterophily: A survey,” 2022.
  58. J. Zhou, G. Cui, S. Hu, Z. Zhang, C. Yang, Z. Liu, L. Wang, C. Li, and M. Sun, “Graph neural networks: A review of methods and applications,” AI Open, vol. 1, pp. 57–81, 2020.
  59. G. Zhu, Z. Zhu, W. Wang, Z. Xu, C. Yuan, and Y. Huang, “Autoac: Towards automated attribute completion for heterogeneous graph neural network,” in ICDE, 2023, pp. 2808–2821.
  60. J. Zhu, Y. Yan, L. Zhao, M. Heimann, L. Akoglu, and D. Koutra, “Beyond homophily in graph neural networks: Current limitations and effective designs,” in NeurIPS, vol. 33, 2020, pp. 7793–7804.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (9)
  1. Tianhao Peng (11 papers)
  2. Wenjun Wu (48 papers)
  3. Haitao Yuan (14 papers)
  4. Zhifeng Bao (28 papers)
  5. Zhao Pengrui (2 papers)
  6. Xin Yu (192 papers)
  7. Xuetao Lin (1 paper)
  8. Yu Liang (57 papers)
  9. Yanjun Pu (4 papers)
Citations (6)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now