Fast geometric learning with symbolic matrices

By Jean Feydy, Joan Alexis Glaunès, Benjamin Charlier, Michael M. Bronstein. 2020

49 citations recorded by the source corpus.

Subfields: Optimization & Training

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Optimization & Training

1. Bruno E. Correia
2. Freyr Sverrisson
3. Tri Dao
4. Christopher R'e
5. Atri Rudra
6. Stefano Ermon
7. Michael M. Bronstein
8. Jean Feydy

Citations in the corpus

• FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness (2022, background): “The KeOps library [8, 28] is a successful example of how reducing memory reads/writes can speed up kernel operations.”
• Fast end-to-end learning on protein surfaces (2020, methodology): “Notably, we implement all core calculations as reductions of symbolic “distance-like” matrices, supported by the recent KeOps library [19] for PyTorch [30]: the high performance routines of this toolbox allow us to design a method which is fully differentiable and an order of magnitude faster and mo”
• P HYSICS - INFORMED DEEP NEURAL NETWORK FOR RIGID - BODY PROTEIN DOCKING (2022, methodology): “In order to increase speed and reduce the memory footprint of our computations, we rely extensively on the KeOps library (Charlier et al., 2021; Feydy et al., 2020).”