GLT has presented the first attempt to accelerate GNN inference. Though promising, GLT encounters robustness and generalization issues when deployed in real-world scenarios. To tackle these issues, we propose the Resilient Graph Lottery Ticket (RGLT) to find more robust and generalizable GLT in GNNs.
If you find this work useful, please cite it as:
@article{wang2023brave, title={Brave the wind and the waves: Discovering robust and generalizable graph lottery tickets}, author={Wang, Kun and Liang, Yuxuan and Li, Xinglin and Li, Guohao and Ghanem, Bernard and Zimmermann, Roger and Yi, Huahui and Zhang, Yudong and Wang, Yang and others}, journal={IEEE Transactions on Pattern Analysis and Machine Intelligence}, year={2023}, publisher={IEEE} }