Get our free extension to see links to code for papers anywhere online!

Chrome logo  Add to Chrome

Firefox logo Add to Firefox

Prospect Pruning: Finding Trainable Weights at Initialization using Meta-Gradients


Feb 16, 2022
Milad Alizadeh, Shyam A. Tailor, Luisa M Zintgraf, Joost van Amersfoort, Sebastian Farquhar, Nicholas Donald Lane, Yarin Gal

Add code


   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Towards Efficient Point Cloud Graph Neural Networks Through Architectural Simplification


Aug 13, 2021
Shyam A. Tailor, René de Jong, Tiago Azevedo, Matthew Mattina, Partha Maji

Add code

* 8 pages. Accepted to the Deep Learning for Geometric Computing Workshop at ICCV 2021 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Adaptive Filters and Aggregator Fusion for Efficient Graph Convolutions


Apr 10, 2021
Shyam A. Tailor, Felix L. Opolka, Pietro Liò, Nicholas D. Lane

Add code

* Short versions to appear at the GNNSys Workshop at MLSys 2021 and the Hardware-Aware Efficient Training Workshop at ICLR 2021 

   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Degree-Quant: Quantization-Aware Training for Graph Neural Networks


Aug 11, 2020
Shyam A. Tailor, Javier Fernandez-Marques, Nicholas D. Lane

Add code


   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email

Are Accelerometers for Activity Recognition a Dead-end?


Jan 30, 2020
Catherine Tong, Shyam A. Tailor, Nicholas D. Lane

Add code


   Access Paper or Ask Questions

  • Share via Twitter
  • Share via Facebook
  • Share via LinkedIn
  • Share via Whatsapp
  • Share via Messenger
  • Share via Email