8 problems

Let $\tau_0^*$ denote the source-model ridge penalty that maximizes transfer benefit, and let SNR denote the signal-to-noise ratio. The relationship observed in the linear-model se…

Let the assumptions in Crefrange{asm:model}{asm:feat} hold. Define $\phi_n=p/n$. Let $k\le n$ be the subsample size and set $\bar{\psi}_n=p/k$. Suppose that $\varphi$ satisfies cer…

Theorem concerns conditions ensuring that the relevant system of equations has a unique and bounded solution. The paper does not specify the technical conditions in the conjectural…

Low-rank prediction conjecture. It is always desirable to use reduced rank multivariate kernel ridge regression when the underlying function $F$ is of low rank.

Let $R_{\rm RF}(f_d, {\boldsymbol X}, {\boldsymbol \Theta}, \lambda)$ denote the prediction risk of random-features ridge regression, with ridge parameter $\lambda>0$, and let the…

Let ridge IV denote the instrumental-variable estimator obtained by adding an $L_2$ penalty to the GMM objective function, with penalty parameter $b[?m\lambdab[0m$. An appropriate…

Let $\mathcal{M}$ be a continuous collection of tuning parameters or estimators, and let Algorithm 5 denote the minimal-penalty procedure considered in the paper. Suppose that a co…

In sketched matrix ridge regression, let the generalization error denote the prediction error on an unseen test sample. Generalization-error conjecture. The generalization error ca…