Skip to content

Latest commit

 

History

History
6 lines (5 loc) · 477 Bytes

README.md

File metadata and controls

6 lines (5 loc) · 477 Bytes

Small-scale experiments with in context learning using transformers

Experiments with in context learning using simple function classes:

  1. Reproducing and extending work in this paper: https://arxiv.org/abs/2208.01066.
  2. Formulated a Bayesian prior-learning hypothesis for the efficacy of in-context learning in transformers.
  3. You can find the project's website here.