4 - Risks from Learned Optimization with Evan Hubinger
AXRP - the AI X-risk Research Podcast
00:00
Do You Know How to Train Look Up Tables?
The key is inducte biases in generalization. There's a reason that we don't train look up tables, and we instead train, yeur lookworks. The direction i expect wi shen learning to continue to go is to sort of develop these sorts of nice, a sort of indoctor ise set up the sort of situations where weso have inductovices that select for simple models.
Transcript
Play full episode
