Improving Graph Generation by Restricting Graph Bandwidth

Nathaniel Lee Diamant, Alex M Tseng, Kangway V. Chuang, Tommaso Biancalani, Gabriele Scalia
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:7939-7959, 2023.

Abstract

Deep graph generative modeling has proven capable of learning the distribution of complex, multi-scale structures characterizing real-world graphs. However, one of the main limitations of existing methods is their large output space, which limits generation scalability and hinders accurate modeling of the underlying distribution. To overcome these limitations, we propose a novel approach that significantly reduces the output space of existing graph generative models. Specifically, starting from the observation that many real-world graphs have low graph bandwidth, we restrict graph bandwidth during training and generation. Our strategy improves both generation scalability and quality without increasing architectural complexity or reducing expressiveness. Our approach is compatible with existing graph generative methods, and we describe its application to both autoregressive and one-shot models. We extensively validate our strategy on synthetic and real datasets, including molecular graphs. Our experiments show that, in addition to improving generation efficiency, our approach consistently improves generation quality and reconstruction accuracy. The implementation is made available.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-diamant23a, title = {Improving Graph Generation by Restricting Graph Bandwidth}, author = {Diamant, Nathaniel Lee and Tseng, Alex M and Chuang, Kangway V. and Biancalani, Tommaso and Scalia, Gabriele}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {7939--7959}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/diamant23a/diamant23a.pdf}, url = {https://proceedings.mlr.press/v202/diamant23a.html}, abstract = {Deep graph generative modeling has proven capable of learning the distribution of complex, multi-scale structures characterizing real-world graphs. However, one of the main limitations of existing methods is their large output space, which limits generation scalability and hinders accurate modeling of the underlying distribution. To overcome these limitations, we propose a novel approach that significantly reduces the output space of existing graph generative models. Specifically, starting from the observation that many real-world graphs have low graph bandwidth, we restrict graph bandwidth during training and generation. Our strategy improves both generation scalability and quality without increasing architectural complexity or reducing expressiveness. Our approach is compatible with existing graph generative methods, and we describe its application to both autoregressive and one-shot models. We extensively validate our strategy on synthetic and real datasets, including molecular graphs. Our experiments show that, in addition to improving generation efficiency, our approach consistently improves generation quality and reconstruction accuracy. The implementation is made available.} }
Endnote
%0 Conference Paper %T Improving Graph Generation by Restricting Graph Bandwidth %A Nathaniel Lee Diamant %A Alex M Tseng %A Kangway V. Chuang %A Tommaso Biancalani %A Gabriele Scalia %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-diamant23a %I PMLR %P 7939--7959 %U https://proceedings.mlr.press/v202/diamant23a.html %V 202 %X Deep graph generative modeling has proven capable of learning the distribution of complex, multi-scale structures characterizing real-world graphs. However, one of the main limitations of existing methods is their large output space, which limits generation scalability and hinders accurate modeling of the underlying distribution. To overcome these limitations, we propose a novel approach that significantly reduces the output space of existing graph generative models. Specifically, starting from the observation that many real-world graphs have low graph bandwidth, we restrict graph bandwidth during training and generation. Our strategy improves both generation scalability and quality without increasing architectural complexity or reducing expressiveness. Our approach is compatible with existing graph generative methods, and we describe its application to both autoregressive and one-shot models. We extensively validate our strategy on synthetic and real datasets, including molecular graphs. Our experiments show that, in addition to improving generation efficiency, our approach consistently improves generation quality and reconstruction accuracy. The implementation is made available.
APA
Diamant, N.L., Tseng, A.M., Chuang, K.V., Biancalani, T. & Scalia, G.. (2023). Improving Graph Generation by Restricting Graph Bandwidth. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:7939-7959 Available from https://proceedings.mlr.press/v202/diamant23a.html.

Related Material