Graph Warp Module: an Auxiliary Module for Boosting the Power of Graph Neural Networks in Molecular Graph Analysis

By Katsuhiko Ishiguro, Shin-ichi Maeda, Masanori Koyama. 2019

34 citations recorded by the source corpus.

Subfields: Graph Learning, Computational Chemistry & Drug Discovery

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Graph Learning

1. Marinka Zitnik
2. Jure Leskovec
3. Weihua Hu
4. Percy Liang
5. William L. Hamilton
6. Bowen Liu
7. Vijay S. Pande
8. Joseph Gomes

In Computational Chemistry & Drug Discovery

1. Jure Leskovec
2. Bowen Liu
3. Joseph Gomes
4. Marinka Zitnik
5. Weihua Hu
6. Percy Liang
7. Vijay S. Pande
8. Guolin Ke

Citations in the corpus

• Strategies for Pre-training Graph Neural Networks (2019, methodology): “In particular, we achieved a ROC-AUC score of 79.9%, while best-performing graph models in Wu et al. (2018), Li et al. (2017), and Ishiguro et al. (2019) had ROC-AUC scores of 76.3%, 77.6%, and 76.2%, respectively.”
• Do Transformers Really Perform Badly for Graph Representation? (2021, methodology): “Besides the superior expressiveness than popular GNNs, we also find an interesting connection between using self-attention and the virtual node heuristic [15, 31, 24, 22].”
• Do Transformers Really Perform Bad for Graph Representation? (2021, methodology): “However, a naive addition of a supernode to a graph can potentially lead to inadvertent over-smoothing of information propagation [23].”