Generalised f-Mean Aggregation for Graph Neural Networks

TL;DR

This paper introduces GenAgg, a flexible aggregation operator for GNNs that encompasses standard aggregators, leading to improved performance by better minimizing information loss during aggregation.

Contribution

The paper proposes GenAgg, a generalized aggregation function for GNNs that can represent standard aggregators with higher accuracy and enhance GNN performance.

Findings

GenAgg outperforms baseline aggregators in representing standard functions.

Using GenAgg as a drop-in replacement boosts GNN task performance.

GenAgg reduces information loss during aggregation.

Abstract

Graph Neural Network (GNN) architectures are defined by their implementations of update and aggregation modules. While many works focus on new ways to parametrise the update modules, the aggregation modules receive comparatively little attention. Because it is difficult to parametrise aggregation functions, currently most methods select a ``standard aggregator'' such as $\mathrm{mean}$, $\mathrm{sum}$, or $\mathrm{max}$. While this selection is often made without any reasoning, it has been shown that the choice in aggregator has a significant impact on performance, and the best choice in aggregator is problem-dependent. Since aggregation is a lossy operation, it is crucial to select the most appropriate aggregator in order to minimise information loss. In this paper, we present GenAgg, a generalised aggregation operator, which parametrises a function space that includes all standard aggregators. In our experiments, we show that GenAgg is able to represent the standard aggregators with much higher accuracy than baseline methods. We also show that using GenAgg as a drop-in replacement for an existing aggregator in a GNN often leads to a significant boost in performance across various tasks.