GraphFM: Improving Large-Scale GNN Training via Feature Momentum

Haiyang Yu, Limei Wang, Bokun Wang, Meng Liu, Tianbao Yang, Shuiwang Ji
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:25684-25701, 2022.

Abstract

Training of graph neural networks (GNNs) for large-scale node classification is challenging. A key difficulty lies in obtaining accurate hidden node representations while avoiding the neighborhood explosion problem. Here, we propose a new technique, named feature momentum (FM), that uses a momentum step to incorporate historical embeddings when updating feature representations. We develop two specific algorithms, known as GraphFM-IB and GraphFM-OB, that consider in-batch and out-of-batch data, respectively. GraphFM-IB applies FM to in-batch sampled data, while GraphFM-OB applies FM to out-of-batch data that are 1-hop neighborhood of in-batch data. We provide a convergence analysis for GraphFM-IB and some theoretical insight for GraphFM-OB. Empirically, we observe that GraphFM-IB can effectively alleviate the neighborhood explosion problem of existing methods. In addition, GraphFM-OB achieves promising performance on multiple large-scale graph datasets.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-yu22g, title = {{G}raph{FM}: Improving Large-Scale {GNN} Training via Feature Momentum}, author = {Yu, Haiyang and Wang, Limei and Wang, Bokun and Liu, Meng and Yang, Tianbao and Ji, Shuiwang}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {25684--25701}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/yu22g/yu22g.pdf}, url = {https://proceedings.mlr.press/v162/yu22g.html}, abstract = {Training of graph neural networks (GNNs) for large-scale node classification is challenging. A key difficulty lies in obtaining accurate hidden node representations while avoiding the neighborhood explosion problem. Here, we propose a new technique, named feature momentum (FM), that uses a momentum step to incorporate historical embeddings when updating feature representations. We develop two specific algorithms, known as GraphFM-IB and GraphFM-OB, that consider in-batch and out-of-batch data, respectively. GraphFM-IB applies FM to in-batch sampled data, while GraphFM-OB applies FM to out-of-batch data that are 1-hop neighborhood of in-batch data. We provide a convergence analysis for GraphFM-IB and some theoretical insight for GraphFM-OB. Empirically, we observe that GraphFM-IB can effectively alleviate the neighborhood explosion problem of existing methods. In addition, GraphFM-OB achieves promising performance on multiple large-scale graph datasets.} }
Endnote
%0 Conference Paper %T GraphFM: Improving Large-Scale GNN Training via Feature Momentum %A Haiyang Yu %A Limei Wang %A Bokun Wang %A Meng Liu %A Tianbao Yang %A Shuiwang Ji %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-yu22g %I PMLR %P 25684--25701 %U https://proceedings.mlr.press/v162/yu22g.html %V 162 %X Training of graph neural networks (GNNs) for large-scale node classification is challenging. A key difficulty lies in obtaining accurate hidden node representations while avoiding the neighborhood explosion problem. Here, we propose a new technique, named feature momentum (FM), that uses a momentum step to incorporate historical embeddings when updating feature representations. We develop two specific algorithms, known as GraphFM-IB and GraphFM-OB, that consider in-batch and out-of-batch data, respectively. GraphFM-IB applies FM to in-batch sampled data, while GraphFM-OB applies FM to out-of-batch data that are 1-hop neighborhood of in-batch data. We provide a convergence analysis for GraphFM-IB and some theoretical insight for GraphFM-OB. Empirically, we observe that GraphFM-IB can effectively alleviate the neighborhood explosion problem of existing methods. In addition, GraphFM-OB achieves promising performance on multiple large-scale graph datasets.
APA
Yu, H., Wang, L., Wang, B., Liu, M., Yang, T. & Ji, S.. (2022). GraphFM: Improving Large-Scale GNN Training via Feature Momentum. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:25684-25701 Available from https://proceedings.mlr.press/v162/yu22g.html.

Related Material