AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
3min chapter
Hattie Zhou: Lottery Tickets and Algorithmic Reasoning in LLMs
The Gradient: Perspectives on AI
Retraining the Later Layers to Improve Transfer Learning Performance?
You mentioned that they were able to improve transfer learning performance by periodically reinitializing their final layer during fine tuning. Can you tell me a little bit about if that kind of got you thinking about any questions and sort of that representation learning perspective on this as well? Yeah, so we also have some similar experiments in the paper that doesn't apply to transfer learning, but does reset the later layers of the network. And I think one of the questions there is, you know, what is causing the improvement after retraining? Is it that you now have a better later layer? Or is it the case that the features that are in the earlier layers are being kind of amplified because they're never
00:00
Transcript
Episode notes
Get the Snipd
podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any
moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share
& Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode