upcarta
  • Sign In
  • Sign Up
  • Explore
  • Search

Joost van Amersfoort

joo.st
1 Follower
community-curated profile

Research Scientist @DeepMind -- PhD from @UniofOxford

Large language model pretraining. Log scale is my default scale.

Used to be a Leo, now a Gemini.

Overview Posts Content Recommendations
Featured content
See All
  • Human in the Loop: Deep Learning without Wasteful Labelling
    by Andreas Kirsch
  • On Feature Collapse and Deep Kernel Learning for Single Forward Pass Uncertainty
    by Yarin Gal
Recent recommendations
See All
  • Beyond Discrete Support in Large-scale Bayesian Deep Learning - OATML
    by Yarin Gal
Recent posts
See All
Joost van Amersfoort @joost_v_amersf · Apr 29, 2020
  • From Twitter

Nice write-up of a simple and effective improvement for training your Bayesian NN using Bayes by Backprop!

Paper Apr 22, 2020
Beyond Discrete Support in Large-scale Bayesian Deep Learning - OATML
by Yarin Gal and 2 others
Post Add to Collection Mark as Completed
Recommended by 1 person
1 mention
Share on Twitter Repost
  • upcarta ©2025
  • Home
  • About
  • Terms
  • Privacy
  • Cookies
  • @upcarta