Improving performance and efficiency of Graph Neural Networks by injective aggregation

作者：

Highlights：

• Theoretical and practical guidance on deriving injective aggregations from graphs.

• Systematical comparison between the injective aggregations.

• Advanced efficiency of combination of pre-aggregation and post-encoding.

摘要

•Theoretical and practical guidance on deriving injective aggregations from graphs.•Systematical comparison between the injective aggregations.•Advanced efficiency of combination of pre-aggregation and post-encoding.

论文关键词：Graph Neural Networks,Aggregation function,Aggregation matrix,Injectivity,Traffic state prediction

论文评审过程：Received 29 October 2021, Revised 30 July 2022, Accepted 3 August 2022, Available online 13 August 2022, Version of Record 23 August 2022.

论文官网地址：https://doi.org/10.1016/j.knosys.2022.109616