RT @instadeepai
๐Set your alarm! Our @iclr_conf Spotlight presentation on Quality-Diversity as a competitive alternative to standard RL methods ๐ starts at 0900 Kigali time - donโt miss it!
๐Head over to Oral 1 Track 3, or log on at the link! #ICLR2023 1/2 https://bit.ly/3LG1fmb
Starting into the weekend with the message that our paper on approximating Riemannian shape manifolds was accepted at #ICLR2023 (especially notable-top-25%) has been great!
A big thanks to my co-authors: Klaus Hildebrandt, Benedikt Wirth, and Martin Rumpf. More details coming soon!
RT @CLNakalembe
Join us in #Kigali! #Rwanda May 5 for the 1st Machine Learning for Remote Sensing Workshop to be hosted at
@iclr_conf. We look forward to your submissions and participation. #ICLR2023
- Website: https://nasaharvest.github.io/ml-for-remote-sensing/iclr2023
- Submission deadline: Feb 3, 2023
RT @ducha_aiki@twitter.com
Identical Initialization: A Universal Approach to Fast and Stable Training of Neural Networks
Anon #ICLR2023 submission
tl;dr: initialize your weights with (almost) identity.
https://openreview.net/forum?id=qpeAhwxTopw
๐ฆ๐: https://twitter.com/ducha_aiki/status/1579786888977997826