Simple GNNs with Low Rank Non-parametric Aggregators
This work addresses the engineering burden in GNN design for researchers and practitioners, but it is incremental as it builds on existing spectral GNN approaches.
The authors tackled the problem of over-engineered GNN architectures for semi-supervised node classification by replacing feature aggregation with a non-parametric learner, resulting in streamlined design and avoidance of hyperparameter complexities, with empirical experiments showing suitability for various graph types.
We revisit recent spectral GNN approaches to semi-supervised node classification (SSNC). We posit that state-of-the-art (SOTA) GNN architectures may be over-engineered for common SSNC benchmark datasets (citation networks, page-page networks, etc.). By replacing feature aggregation with a non-parametric learner we are able to streamline the GNN design process and avoid many of the engineering complexities associated with SOTA hyperparameter selection (GNN depth, non-linearity choice, feature dropout probability, etc.). Our empirical experiments suggest conventional methods such as non-parametric regression are well suited for semi-supervised learning on sparse, directed networks and a variety of other graph types commonly found in SSNC benchmarks. Additionally, we bring attention to recent changes in evaluation conventions for SSNC benchmarking and how this may have partially contributed to rising performances over time.