Graph Edit Distance (GED)

Graph Edit Distance (GED) #

Graph Edit Distance measures the minimum cost of transformations needed to change one graph into another. It has applications in pattern matching and graph similarity computation.

Recent Literature #

1. SimGNN - A Neural Network Approach to Fast Graph Similarity Computation WSDM, 2019. paper, code

   Bai, Yunsheng and Ding, Hao and Bian, Song and Chen, Ting and Sun, Yizhou and Wang, Wei

2. Graph Matching Networks for Learning the Similarity of Graph Structured Objects ICML, 2019. paper, code

   Li, Yujia and Gu, Chenjie and Dullien, Thomas and Vinyals, Oriol and Kohli, Pushmeet

3. Convolutional Embedding for Edit Distance SIGIR, 2020. paper, code

   Dai, Xinyan and Yan, Xiao and Zhou, Kaiwen and Wang, Yuxuan and Yang, Han and Cheng, James

4. Learning-Based Efficient Graph Similarity Computation via Multi-Scale Convolutional Set Matching AAAI, 2020. paper, code

   Bai, Yunsheng and Ding, Hao and Gu, Ken and Sun, Yizhou and Wang, Wei

5. ⭐A Bi-Level Framework for Learning to Solve Combinatorial Optimization on Graphs NeurIPS, 2021. paper, code

   Wang, Runzhong and Hua, Zhigang and Liu, Gan and Zhang, Jiayi and Yan, Junchi and Qi, Feng and Yang, Shuang and Zhou, Jun and Yang, Xiaokang

6. ⭐Combinatorial Learning of Graph Edit Distance via Dynamic Embedding. CVPR, 2021. paper, code

   Wang, Runzhong and Zhang, Tianqi and Yu, Tianshu and Yan, Junchi and Yang, Xiaokang.