Kajetan Dymkiewicz
@kdymkiewicz.bsky.social
PhD student @ Language Technology Lab, University of Cambridge. Making GPUs go brrrr
Finally got this beautiful piece from @francois.fleuret.org
January 10, 2025 at 10:44 AM
Finally got this beautiful piece from @francois.fleuret.org
Reposted by Kajetan Dymkiewicz
Griffin Scholars (very smart Ukrainian undergraduates from top unis) are looking for research internships in maths or CS.
(many of them can't return to Ukraine for the summer).
If you can host them (we may help with funding), please fill out this form:
ferenchuszar1.typeform.com/to/WjXursg3
(many of them can't return to Ukraine for the summer).
If you can host them (we may help with funding), please fill out this form:
ferenchuszar1.typeform.com/to/WjXursg3
December 12, 2024 at 3:12 PM
Griffin Scholars (very smart Ukrainian undergraduates from top unis) are looking for research internships in maths or CS.
(many of them can't return to Ukraine for the summer).
If you can host them (we may help with funding), please fill out this form:
ferenchuszar1.typeform.com/to/WjXursg3
(many of them can't return to Ukraine for the summer).
If you can host them (we may help with funding), please fill out this form:
ferenchuszar1.typeform.com/to/WjXursg3
Reposted by Kajetan Dymkiewicz
1) AIs are trained as black boxes, making it hard to understand or control their behavior. This is bad for safety! But what is an alternative? Our idea: train structure into a neural network by configuring which components update on different tasks. We call it "gradient routing."
December 6, 2024 at 10:14 PM
1) AIs are trained as black boxes, making it hard to understand or control their behavior. This is bad for safety! But what is an alternative? Our idea: train structure into a neural network by configuring which components update on different tasks. We call it "gradient routing."