Token and Head Adaptive Transformers for Efficient Natural Language Processing

no code implementations • COLING 2022 • Chonghan Lee, Md Fahim Faysal Khan, Rita Brugarolas Brufau, Ke Ding, Vijaykrishnan Narayanan

While pre-trained language models like BERT have achieved impressive results on various natural language processing tasks, deploying them on resource-restricted devices is challenging due to their intensive computational cost and memory footprint.

BatchGNN: Efficient CPU-Based Distributed GNN Training on Very Large Graphs

no code implementations • 23 Jun 2023 • Loc Hoang, Rita Brugarolas Brufau, Ke Ding, Bo Wu

We present BatchGNN, a distributed CPU system that showcases techniques that can be used to efficiently train GNNs on terabyte-sized graphs.

graph partitioning

Learning Reduced-Order Models for Cardiovascular Simulations with Graph Neural Networks

1 code implementation • 13 Mar 2023 • Luca Pegolotti, Martin R. Pfaller, Natalia L. Rubio, Ke Ding, Rita Brugarolas Brufau, Eric Darve, Alison L. Marsden

Our numerical results demonstrate the accuracy and generalizability of our method in physiological geometries comprising a variety of anatomies and boundary conditions.