normative
Overfit Then Compress as Path to Generalization
A dataset-learning strategy is to first over-parameterize and overfit to prove a learning relationship exists, then compress to the smallest parameter count, since compression is what produces generalization.
normativepending
Speaker
Neil MovvaEvidence Quote
“Once you can overfit, then you can compress, and the compression is how you get generalization.”
Created: 11/17/2026, 6:43:49 AM
Your notes
Loading notes...