# Making Graph Neural Networks Worth It for Low-Data Molecular Machine Learning

arXiv: https://arxiv.org/abs/2011.12203 · HTML: https://mlsanity.com/paper/2011.12203/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 24 Nov 2020.

## Built on by (most cited first, 5 of 5)

- [Ab-Initio Potential Energy Surfaces by Pairing GNNs with Neural Wave Functions](https://mlsanity.com/paper/2110.05064/) (2021)
- [Molecular Topological Profile (MOLTOP) -- Simple and Strong Baseline for Molecular Graph Classification](https://mlsanity.com/paper/2407.12136/) (2024)
- [Scalable Geometric Deep Learning on Molecular Graphs](https://mlsanity.com/paper/2112.03364/) (2021)
- [Embracing assay heterogeneity with neural processes for markedly improved bioactivity predictions](https://mlsanity.com/paper/2308.09086/) (2023)
- [Bringing Atomistic Deep Learning to Prime Time](https://mlsanity.com/paper/2112.04977/) (2021)

## Abstract (arXiv)

Graph neural networks have become very popular for machine learning on molecules due to the expressive power of their learnt representations. However, molecular machine learning is a classically low-data regime and it isn't clear that graph neural networks can avoid overfitting in low-resource settings. In contrast, fingerprint methods are the traditional standard for low-data environments due to their reduced number of parameters and manually engineered features. In this work, we investigate whether graph neural networks are competitive in small data settings compared to the parametrically 'cheaper' alternative of fingerprint methods. When we find that they are not, we explore pretraining and the meta-learning method MAML (and variants FO-MAML and ANIL) for improving graph neural network performance by transfer learning from related tasks. We find that MAML and FO-MAML do enable the graph neural network to outperform models based on fingerprints, providing a path to using graph neural networks even in settings with severely restricted data availability. In contrast to previous work, we find ANIL performs worse that other meta-learning approaches in this molecule setting. Our results suggest two reasons: molecular machine learning tasks may require significant task-specific adaptation, and distribution shifts in test tasks relative to train tasks may contribute to worse ANIL performance.