Sequential Signal Mixing Aggregation for Message Passing Graph Neural Networks

TL;DR

This paper introduces Sequential Signal Mixing Aggregation (SSMA), a novel method for message passing graph neural networks that enhances feature mixing among neighbors, leading to significant performance improvements across multiple benchmarks.

Contribution

The paper proposes SSMA, a new aggregation technique that treats neighbor features as signals and convolves them sequentially, addressing limitations of sum-based aggregators.

Findings

SSMA improves feature mixing in MPGNNs.

SSMA achieves state-of-the-art results on various benchmarks.

Extensive experiments validate the effectiveness of SSMA.

Abstract

Message Passing Graph Neural Networks (MPGNNs) have emerged as the preferred method for modeling complex interactions across diverse graph entities. While the theory of such models is well understood, their aggregation module has not received sufficient attention. Sum-based aggregators have solid theoretical foundations regarding their separation capabilities. However, practitioners often prefer using more complex aggregations and mixtures of diverse aggregations. In this work, we unveil a possible explanation for this gap. We claim that sum-based aggregators fail to "mix" features belonging to distinct neighbors, preventing them from succeeding at downstream tasks. To this end, we introduce Sequential Signal Mixing Aggregation (SSMA), a novel plug-and-play aggregation for MPGNNs. SSMA treats the neighbor features as 2D discrete signals and sequentially convolves them, inherently enhancing the ability to mix features attributed to distinct neighbors. By performing extensive experiments, we show that when combining SSMA with well-established MPGNN architectures, we achieve substantial performance gains across various benchmarks, achieving new state-of-the-art results in many settings. We published our code at \url{https://almogdavid.github.io/SSMA/}