Elastic Graph Neural Networks

07/05/2021
by   Xiaorui Liu, et al.
0

While many existing graph neural networks (GNNs) have been proven to perform ℓ_2-based graph smoothing that enforces smoothness globally, in this work we aim to further enhance the local smoothness adaptivity of GNNs via ℓ_1-based graph smoothing. As a result, we introduce a family of GNNs (Elastic GNNs) based on ℓ_1 and ℓ_2-based graph smoothing. In particular, we propose a novel and general message passing scheme into GNNs. This message passing algorithm is not only friendly to back-propagation training but also achieves the desired smoothing properties with a theoretical convergence guarantee. Experiments on semi-supervised learning tasks demonstrate that the proposed Elastic GNNs obtain better adaptivity on benchmark datasets and are significantly robust to graph adversarial attacks. The implementation of Elastic GNNs is available at <https://github.com/lxiaorui/ElasticGNN>.

READ FULL TEXT
research
05/30/2020

Understanding the Message Passing in Graph Neural Networks via Power Iteration

The mechanism of message passing in graph neural networks(GNNs) is still...
research
02/08/2022

Boosting Graph Neural Networks by Injecting Pooling in Message Passing

There has been tremendous success in the field of graph neural networks ...
research
11/08/2022

Improving Graph Neural Networks at Scale: Combining Approximate PageRank and CoreRank

Graph Neural Networks (GNNs) have achieved great successes in many learn...
research
01/05/2023

Randomized Message-Interception Smoothing: Gray-box Certificates for Graph Neural Networks

Randomized smoothing is one of the most promising frameworks for certify...
research
06/03/2023

Can Directed Graph Neural Networks be Adversarially Robust?

The existing research on robust Graph Neural Networks (GNNs) fails to ac...
research
06/22/2023

StrainNet: Predicting crystal structure elastic properties using SE(3)-equivariant graph neural networks

Accurately predicting the elastic properties of crystalline solids is vi...
research
06/15/2020

Optimization and Generalization Analysis of Transduction through Gradient Boosting and Application to Multi-scale Graph Neural Networks

It is known that the current graph neural networks (GNNs) are difficult ...

Please sign up or login with your details

Forgot password? Click here to reset