Designing the Topology of Graph Neural Networks: A Novel Feature Fusion Perspective

by   Lanning Wei, et al.

In recent years, Graph Neural Networks (GNNs) have shown superior performance on diverse real-world applications. To improve the model capacity, besides designing aggregation operations, GNN topology design is also very important. In general, there are two mainstream GNN topology design manners. The first one is to stack aggregation operations to obtain the higher-level features but easily got performance drop as the network goes deeper. Secondly, the multiple aggregation operations are utilized in each layer which provides adequate and independent feature extraction stage on local neighbors while are costly to obtain the higher-level information. To enjoy the benefits while alleviating the corresponding deficiencies of these two manners, we learn to design the topology of GNNs in a novel feature fusion perspective which is dubbed F^2GNN. To be specific, we provide a feature fusion perspective in designing GNN topology and propose a novel framework to unify the existing topology designs with feature selection and fusion strategies. Then we develop a neural architecture search method on top of the unified framework which contains a set of selection and fusion operations in the search space and an improved differentiable search algorithm. The performance gains on eight real-world datasets demonstrate the effectiveness of F^2GNN. We further conduct experiments to show that F^2GNN can improve the model capacity while alleviating the deficiencies of existing GNN topology design manners, especially alleviating the over-smoothing problem, by utilizing different levels of features adaptively.


page 1

page 7

page 8

page 13


Enhancing Intra-class Information Extraction for Heterophilous Graphs: One Neural Architecture Search Approach

In recent years, Graph Neural Networks (GNNs) have been popular in graph...

Learn Layer-wise Connections in Graph Neural Networks

In recent years, Graph Neural Networks (GNNs) have shown superior perfor...

Semi-Equivariant GNN Architectures for Jet Tagging

Composing Graph Neural Networks (GNNs) of operations that respect physic...

Pooling Architecture Search for Graph Classification

Graph classification is an important problem with applications across ma...

Graph Property Prediction on Open Graph Benchmark: A Winning Solution by Graph Neural Architecture Search

Aiming at two molecular graph datasets and one protein association subgr...

TFE-GNN: A Temporal Fusion Encoder Using Graph Neural Networks for Fine-grained Encrypted Traffic Classification

Encrypted traffic classification is receiving widespread attention from ...

GraTO: Graph Neural Network Framework Tackling Over-smoothing with Neural Architecture Search

Current Graph Neural Networks (GNNs) suffer from the over-smoothing prob...

Please sign up or login with your details

Forgot password? Click here to reset