`

Timezone: »

 
Scalable Geometric Deep Learning on Molecular Graphs
Nathan Frey · Siddharth Samsi · Lin Li · Connor Coley
Event URL: https://openreview.net/forum?id=oeq0YQYn8Dv »

Deep learning in molecular and materials sciences is limited by the lack of integration between applied science, artificial intelligence, and high-performance computing. Bottlenecks with respect to the amount of training data, the size and complexity of model architectures, and the scale of the compute infrastructure are all key factors limiting the scaling of deep learning for molecules and materials. Here, we present LitMatter, a lightweight framework for scaling molecular deep learning methods. We train four graph neural network architectures on over 400 GPUs and investigate the scaling behavior of these methods. Depending on the model architecture, training time speedups up to 60x are seen. Empirical neural scaling relations quantify the model-dependent scaling and enable optimal compute resource allocation and the identification of scalable molecular geometric deep learning model implementations.

Author Information

Nathan Frey (Massachusetts Institute of Technology)

I am a theoretical materials physicist, National Defense Science & Engineering Graduate fellow, and PhD candidate in Materials Science & Engineering at the University of Pennsylvania in Philadelphia. I use multiscale modeling and computational techniques, including machine learning, to discover and design new materials for next-generation information processing platforms and sustainable energy applications.

Siddharth Samsi (MIT Lincoln Laboratory)
Lin Li (MIT Lincoln Laboratory)
Connor Coley (MIT)

More from the Same Authors