Grokking vs learning: same features, different encodings
Grokking typically achieves similar losses to ordinary, ‘steady’, learning. This work asks whether these different learning paths lead to fundamental differences in the learned models. To do so, we compare the features, compressibility, and learning dynamics of models trained via each path in two co...
Sparad:
| Huvudupphov: | , , , , , , |
|---|---|
| Materialtyp: | Artigo |
| Språk: | Inglês |
| Utgiven: |
IOP Publishing
2026-01-01
|
| Serie: | Machine Learning: Science and Technology |
| Ämnen: | |
| Länkar: | https://doi.org/10.1088/2632-2153/ae67d0 |
| Taggar: |
Inga taggar, Lägg till första taggen!
|