← Search

Vai Suliafu

2 accepted papers

2021

FMMformer: Efficient and Flexible Transformer via Decomposed Near-field and Far-field Attention

NeurIPS 2021poster

We propose FMMformers, a class of efficient and flexible transformers inspired by the celebrated fast multipole method (FMM) for accelerating interacting particle simulation. FMM decomposes particle-particle interaction into near-field and far-field components and then performs direct and coarse-gra…

Cited by 33SourcePDFScholar
2021

Heavy Ball Neural Ordinary Differential Equations

NeurIPS 2021poster

We propose heavy ball neural ordinary differential equations (HBNODEs), leveraging the continuous limit of the classical momentum accelerated gradient descent, to improve neural ODEs (NODEs) training and inference. HBNODEs have two properties that imply practical advantages over NODEs: (i) The adjoi…