Grafik Sinir Ağları Üzerine Bir İnceleme

Yıl 2024, , 105 - 138, 07.05.2024

Hamza Talha Gümüş , Can Eyüpoğlu

https://doi.org/10.17134/khosbd.1284174

Öz

Grafik sinir ağları (Graph Neural Networks-GNN), yapay sinir ağları (Artificial Neural Networks-ANN) ailesine mensup ve grafikler üzerinden bilgi çıkarımı işlemi gerçekleştiren bir derin öğrenme yöntemidir. Bilinen ilk modeller GNN tabanlı oluşturulsa da evrişimli ağların kullanımının yaygınlaşması ile grafik evrişimli ağ (Graph Convolutional Network-GCN) modeli de popülerlik kazanmıştır. Bu durum GNN’lerin gelişmesine katkı sağlarken, iki ağ tabanında yeni modeller oluşturulmasına öncül olmuştur. Bu çalışmada GNN ve GCN modelleri tabanında oluşturulmuş 65 alt model ve bu modellerin etkilediği diğer modeller incelenmiştir.

Anahtar Kelimeler

Grafik Sinir Ağları, Yapay Sinir Ağları, Derin Öğrenme, Grafik Evrişimli Ağlar

Teşekkür

Bu çalışma Milli Savunma Üniversitesi, Atatürk Stratejik Araştırmalar ve Lisansüstü Eğitim Enstitüsü, Bilgisayar Mühendisliği Anabilim Dalı, Siber Güvenlik Tezsiz Yüksek Lisans Programına bağlı olarak yürütülen “Siber Güvenlikte Grafik Sinir Ağları” adlı dönem projesinin bir bölümüdür.

A Review on Graph Neural Networks

Öz

Graph neural networks (GNN) are a deep learning method that belongs to the family of artificial neural networks (ANN) and performs information extraction over graphs. Although the first known models were created based on GNN, the Graph Convolutional Network (GCN) model has gained popularity with the widespread use of convolutional networks. While this has contributed to the development of GNNs, it led to the creation of new models on two network bases. In this study, 65 sub-models created on the basis of GNN and GCN models and other models affected by these models were examined.

Anahtar Kelimeler

Graph Neural Networks, Artificial Neural Networks, Deep Learning, Graph Convolutional Networks

Kaynakça

• [1] F. Scarselli, M. Gori, A. C. Tsoi, M. Hagenbuchner, and G. Monfardini, “The Graph Neural Network Model,” IEEE Transactions on Neural Networks, vol. 20, no. 1, pp. 61-80, Jan. 2009, doi: 10.1109/TNN.2008.2005605.
• [2] paperswithcode.com, “Papers with Code - An Overview of Graph Models,” https://paperswithcode.com/methods/category/graph-models (accessed May 5, 2023).
• [3] T. N. Kipf and M. Welling, “Semi-supervised classification with graph convolutional networks,” 5th International Conference on Learning Representations, ICLR 2017, arXiv preprint arXiv:1609.02907 (2016).
• [4] W. L. Chiang, X. Liu, S. Si, Y. Li, S. Bengio, and C. J. Hsieh, “Cluster-GCN: An Efficient Algorithm for Training Deep and Large Graph Convolutional Networks,” Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (KDD '19). Association for Computing Machinery, New York, NY, USA, 2019, pp. 257-266, doi: 10.1145/3292500.3330925.
• [5] J. Chen, J. Zhu, and L. Song, “Stochastic training of graph convolutional networks with variance reduction,” Proceedings of the 35th International Conference on Machine Learning, arXiv preprint arXiv:1710.10568 (2017).
• [6] Y. Shi, Z. Huang, S. Feng, W. Wang, and Y. Sun, “Masked label prediction: Unified message passing model for semi-supervised classification,” arXiv, arXiv:2009.03509 (2020).
• [7] L. Shi, Y. Zhang, J. Cheng, and H. Lu, “Two-Stream Adaptive Graph Convolutional Networks for Skeleton-Based Action Recognition,” 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA, 2019, pp. 12018-12027, doi: 10.1109/CVPR.2019.01230.
• [8] Jia et al., “Predicting Path Failure In Time-Evolving Graphs,” Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, July 2019, pp. 1279-1289, doi: 10.1145/3292500.3330847.
• [9] B. Chen, R. Barzilay, and T. Jaakkola, “Path-Augmented Graph Transformer Network,” arXiv, May 2019, doi: 10.48550/arxiv.1905.12712.
• [10] L. Yao, C. Mao, and Y. Luo, “Graph Convolutional Networks for Text Classification,” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 33, no. 01, pp. 7370-7377, Jul. 2019, pp. 7370-7377, doi: 10.1609/aaai.v33i01.33017370. [11] X. Wang, Y. Ye, and A. Gupta, “Zero-Shot Recognition via Semantic Embeddings and Knowledge Graphs,” 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, 2018, pp. 6857-6866, doi: 10.1109/CVPR.2018.00717. [12] L. Zhao et al., “T-GCN: A Temporal Graph Convolutional Network for Traffic Prediction,” IEEE Transactions on Intelligent Transportation Systems, vol. 21, no. 9, pp. 3848-3858, Sept. 2020, doi: 10.1109/TITS.2019.2935152. [13] M. Xu, C. Zhao, D. S. Rojas, A. Thabet, and B. Ghanem, “G-TAD: Sub-Graph Localization for Temporal Action Detection,” 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, 2020, pp. 10153-10162, doi: 10.1109/CVPR42600.2020.01017.