Systematic experimental analysis of the role of feature learning in continual learning using scaling limits!
Posts by Jacopo Graldi
π‘ Check out our paper here openreview.net/pdf?id=TYPBY...
π
Or come to our poster on December 14th, Rooms 109/110 at 11.50-13.30 @neuripsconf.bsky.social
w/ @lorenzonoci.bsky.social
3/3
β Could increasing model scale solve Catastrophic Forgetting in Continual Learning?
β Is feature learning helpful or harmful for Continual Learning?
Our results dive into these questions with surprising insights!
π Curious to learn more? π
2/3
π Excited to be in #Vancouver next week for #NeurIPS to present results from my Masterβs Thesis at the Scalable Continual Learning Workshop on December 14th!
π Our work investigates the role of scale and training regimes in Continual Learning.
What did we find? π
1/3
@support.bsky.team
Hey @bsky.app! Is there any way to include reposts in lists used as feeds? Currently, only direct posts of list members are visualized!