Efficient and Minimax Optimal In-context Nonparametric Regression with Transformers

Michelle Ching, Ioana Popescu, Nico Smith +3

We study in-context learning for nonparametric regression with $\hat{I}\pm$-Hölder smooth regression functions, for some $\hat{I}\pm >0$. We prove that, with $n$ in-context examples and $d$-dimens…