GCN-SE: Attention as Explainability for Node Classification in Dynamic Graphs

10/11/2021
by   Yucai Fan, et al.
1

Graph Convolutional Networks (GCNs) are a popular method from graph representation learning that have proved effective for tasks like node classification tasks. Although typical GCN models focus on classifying nodes within a static graph, several recent variants propose node classification in dynamic graphs whose topologies and node attributes change over time, e.g., social networks with dynamic relationships, or literature citation networks with changing co-authorships. These works, however, do not fully address the challenge of flexibly assigning different importance to snapshots of the graph at different times, which depending on the graph dynamics may have more or less predictive power on the labels. We address this challenge by proposing a new method, GCN-SE, that attaches a set of learnable attention weights to graph snapshots at different times, inspired by Squeeze and Excitation Net (SE-Net). We show that GCN-SE outperforms previously proposed node classification methods on a variety of graph datasets. To verify the effectiveness of the attention weight in determining the importance of different graph snapshots, we adapt perturbation-based methods from the field of explainable machine learning to graphical settings and evaluate the correlation between the attention weights learned by GCN-SE and the importance of different snapshots over time. These experiments demonstrate that GCN-SE can in fact identify different snapshots' predictive power for dynamic node classification.

READ FULL TEXT
research
02/17/2020

Unifying Graph Convolutional Neural Networks and Label Propagation

Label Propagation (LPA) and Graph Convolutional Neural Networks (GCN) ar...
research
07/10/2022

Scalable Privacy-enhanced Benchmark Graph Generative Model for Graph Convolutional Networks

A surge of interest in Graph Convolutional Networks (GCN) has produced t...
research
11/30/2018

Graph Node-Feature Convolution for Representation Learning

Graph convolutional network (GCN) is an emerging neural network approach...
research
05/22/2019

Simulation and Augmentation of Social Networks for Building Deep Learning Models

A limitation of the Graph Convolutional Networks (GCN) is that it assume...
research
09/24/2019

Semi-supervised classification on graphs using explicit diffusion dynamics

Classification tasks based on feature vectors can be significantly impro...
research
06/29/2020

Approximating Network Centrality Measures Using Node Embedding and Machine Learning

Analyzing and extracting useful information from real-world complex netw...
research
09/24/2019

Layerwise Relevance Visualization in Convolutional Text Graph Classifiers

Representations in the hidden layers of Deep Neural Networks (DNN) are o...

Please sign up or login with your details

Forgot password? Click here to reset