Improved Search in Graph AutoML: Expansion and Dynamic Prioritization in the Search Space for Enhanced Efficiency
https://doi.org/10.15514/ISPRAS-2025-37(2)-8
Abstract
This paper explores methods for enhancing the automated architecture search process for graph neural networks. We propose a novel approach that dynamically selects a priority direction within the search space, improving the efficiency and quality of the discovered architectures. Another proposed approach expands the search space by allowing combinations of different types of graph convolutional layers. The primary focus is on maximizing the quality of architectures within the expanded search space while maintaining a fixed search budget in terms of the number of models. Our experiments are conducted on datasets from citation networks, chemical molecules, and shopping graph domains. The experimental results show that the proposed approach enables the discovery of more effective and higher-quality models without increasing computational resources, demonstrating high potential for automating solutions to real-world graph data analysis tasks.
About the Authors
Fedor Mikhailovich BALABANOV
Ivannikov Institute for System Programming of the Russian Academy of Sciences,
Lomonosov Moscow State University
Russian Federation
Russian Federation
Laboratory assistant in the Ivannikov Institute for System Programming of the Russian Academy of Sciences, student at the Moscow State University. His research interests are graph neural networks, AutoML.
Kirill Sergeevich LUKYANOV
Ivannikov Institute for System Programming of the Russian Academy of Sciences,
Moscow Institute of Physics and Technology (National Research University),
Research Center for Trusted Artificial Intelligence ISP RAS
Russian Federation
Russian Federation
Researcher at the Center for Trusted Artificial Intelligence of the Ivannikov Institute for System Programming of the Russian Academy of Sciences; postgraduate student at Moscow Institute of Physics and Technology. Research interests: trustworthy artificial intelligence with a particular focus on studies at the intersection of multiple trust criteria being ensured simultaneously (e.g., achieving both interpretability and robustness of AI models), AutoML, data domains – graphs, images, time series, tabular data.
References
1. Xiao Li, Li Sun, Mengjie Ling, Yan Peng A survey of graph neural network based recommendation in social networks. Neurocomputing, 2023, 126441, https://doi.org/10.1016/j.neucom.2023.126441.
2. Fan, Wenqi, Ma, Yao, Li, Qing, He, Yuan, Zhao, Eric, Tang, Jiliang, Yin, Dawei Graph Neural Networks for Social Recommendation. 2019, 417–426, 10.1145/3308558.3313488.
3. Park, Jaehong, Shim, Youngseon, Lee, Franklin, Rammohan, Aravind, Goyal, Sushmit, Shim, Munbo, Jeong, Changwook, Kim, Dae Sin Prediction and Interpretation of Polymer Properties Using the Graph Convolutional Network. ACS Polymers Au, 2022, 213-222, 10.1021/acspolymersau.1c00050.
4. Pietro Bongini, Monica Bianchini, Franco Scarselli Molecular generative Graph Neural Networks for Drug Discovery. Neurocomputing, 2021, 242-252, https://doi.org/10.1016/j.neucom.2021.04.039.
5. Fan Zhou, Xovee Xu, Ce Li, Goce Trajcevski, Ting Zhong, Kunpeng Zhang A Heterogeneous Dynamical Graph Neural Networks Approach to Quantify Scientific Impact. 2020.
6. Cummings, Daniel, Nassar, Marcel Structured Citation Trend Prediction Using Graph Neural Networks. 2020, 3897-3901, 10.1109/ICASSP40776.2020.9054769.
7. Xin He, Kaiyong Zhao, Xiaowen Chu AutoML: A survey of the state-of-the-art. Knowledge-Based Systems, 2021, 106622, https://doi.org/10.1016/j.knosys.2020.106622.
8. Rashidi, Hooman H., Tran, Nam, Albahra, Samer, Dang, Luke T. Machine learning in health care and laboratory medicine: General overview of supervised learning and Auto-ML. International Journal of Laboratory Hematology, 2021, 15-22, https://doi.org/10.1111/ijlh.13537.
9. Zoph, Barret, Le, Quoc Neural Architecture Search with Reinforcement Learning. 2016.
10. Hochreiter, S Long Short-term Memory. Neural Computation MIT-Press, 1997.
11. Jie Zhou, Ganqu Cui, Shengding Hu, Zhengyan Zhang, Cheng Yang, Zhiyuan Liu, Lifeng Wang, Changcheng Li, Maosong Sun Graph neural networks: A review of methods and applications. AI Open, 2020, 57-81, https://doi.org/10.1016/j.aiopen.2021.01.001.
12. Gao, Yang, Yang, Hong, Zhang, Peng, Zhou, Chuan GraphNAS: Graph Neural Architecture Search with Reinforcement Learning. 2019.
13. Jiaxuan You, Rex Ying, Jure Leskovec Design Space for Graph Neural Networks. 2021.
14. Zhao, Huan, Wei, Lanning, Yao, Quanming Simplifying Architecture Search for Graph Neural Network. 2020.
15. Zhao, Huan, Yao, Quanming, Tu, Weiwei Search to aggregate neighborhood for graph neural network. 2021.
16. Cai, Rongshen and Tao, Qian and Tang, Yufei and Shi, Min ALGNN: Auto-Designed Lightweight Graph Neural Network. 2021, 500--512.
17. Pytorch library, https://pytorch.org
18. Pytorch Geometric library, https://pytorch-geometric.readthedocs.io
19. Thomas N. Kipf, Max Welling Semi-Supervised Classification with Graph Convolutional Networks. 2017.
20. William L. Hamilton, Rex Ying, Jure Leskovec Inductive Representation Learning on Large Graphs. 2018.
21. Petar Veličković, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Liò, Yoshua Bengio Graph Attention Networks. 2018.
22. Jian Du, Shanghang Zhang, Guanhang Wu, Jose M. F. Moura, Soummya Kar Topology Adaptive Graph Convolutional Networks. 2018.
23. Keyulu Xu, Weihua Hu, Jure Leskovec, Stefanie Jegelka How Powerful are Graph Neural Networks? 2019.
24. Felix Wu, Tianyi Zhang, Amauri Holanda de Souza Jr. au2, Christopher Fifty, Tao Yu, Kilian Q. Weinberger Simplifying Graph Convolutional Networks. 2019.
25. Hao Zhu, Piotr Koniusz Simple Spectral Graph Convolution. 2021.
26. Federico Monti, Davide Boscaini, Jonathan Masci, Emanuele Rodolà, Jan Svoboda, Michael M. Bronstein Geometric deep learning on graphs and manifolds using mixture model CNNs. 2016.
Review
For citations:
BALABANOV F.M., LUKYANOV K.S. Improved Search in Graph AutoML: Expansion and Dynamic Prioritization in the Search Space for Enhanced Efficiency. Proceedings of the Institute for System Programming of the RAS (Proceedings of ISP RAS). 2025;37(2):115-128. (In Russ.) https://doi.org/10.15514/ISPRAS-2025-37(2)-8
Email this article 