# Training speedups via batching for geometric learning: an analysis of static and dynamic algorithms

arXiv: https://arxiv.org/abs/2502.00944 · HTML: https://mlsanity.com/paper/2502.00944/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 41 references that are in the map.
- First posted on arXiv 2 Feb 2025.

## Builds on (most cited first, 14 of 14)

- [Semi-Supervised Classification with Graph Convolutional Networks](https://mlsanity.com/paper/1609.02907/) (2016)
- [How Powerful are Graph Neural Networks?](https://mlsanity.com/paper/1810.00826/) (2018)
- [Fast Graph Representation Learning with PyTorch Geometric](https://mlsanity.com/paper/1903.02428/) (2019)
- [Learning to Simulate Complex Physics with Graph Networks](https://mlsanity.com/paper/2002.09405/) (2020)
- [Carbon Emissions and Large Neural Network Training](https://mlsanity.com/paper/2104.10350/) (2021)
- [Equivariant message passing for the prediction of tensorial properties and molecular spectra](https://mlsanity.com/paper/2102.03150/) (2021)
- [Orb: A Fast, Scalable Neural Network Potential](https://mlsanity.com/paper/2410.22570/) (2024)
- [Neural Message Passing with Edge Updates for Predicting Properties of Molecules and Materials](https://mlsanity.com/paper/1806.03146/) (2018)
- [TF-GNN: Graph Neural Networks in TensorFlow](https://mlsanity.com/paper/2207.03522/) (2022)
- [Learned Force Fields Are Ready For Ground State Catalyst Discovery](https://mlsanity.com/paper/2209.12466/) (2022)
- [Neural Architecture Search for Energy Efficient Always-on Audio Models](https://mlsanity.com/paper/2202.05397/) (2022)
- [Batch Size Influence on Performance of Graphic and Tensor Processing Units during Training and Inference Phases](https://mlsanity.com/paper/1812.11731/) (2018)
- [Band-gap regression with architecture-optimized message-passing neural networks](https://mlsanity.com/paper/2309.06348/) (2023)
- [How big is Big Data?](https://mlsanity.com/paper/2405.11404/) (2024)

## Built on by (most cited first, 2 of 2)

- [Agentic AI Scientists Are Not Built For Autonomous Scientific Discovery](https://mlsanity.com/paper/2605.08956/) (2026)
- [Atompack: A Storage and Distribution Layer for Read-Heavy Atomistic ML Training Datasets](https://mlsanity.com/paper/2606.29975/) (2026)

## Abstract (arXiv)

Graph neural networks (GNN) have shown promising results for several domains such as materials science, chemistry, and the social sciences. GNN models often contain millions of parameters, and like other neural network (NN) models, are often fed only a fraction of the graphs that make up the training dataset in batches to update model parameters. The effect of batching algorithms on training time and model performance has been thoroughly explored for NNs but not yet for GNNs. We analyze two different batching algorithms for graph-based models, namely static and dynamic batching for two datasets, the QM9 dataset of small molecules and the AFLOW materials database. Our experiments show that changing the batching algorithm can provide up to a 2.7x speedup, but the fastest algorithm depends on the data, model, batch size, hardware, and number of training steps run. Experiments show that for a select number of combinations of batch size, dataset, and model, significant differences in model learning metrics are observed between static and dynamic batching algorithms.