AI Paper Atlas

Research map / Learning theory and algorithms / Kernels and regression theory

Out-of-Distribution Generalization in Kernel Regression

Abdulkadir Canatar, Blake Bordelon, Cengiz Pehlevan

NeurIPS 2021 · Learning theory and algorithms › Kernels and regression theory · approach: High-dimensional asymptotic generalization · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Optimize test measure by gradient descent on theoretical E_g (beneficial shift) versus matched train/testNTK regression, depth-3 ReLU NTK · MNIST 8s and 9s, M=1000generalization errorImproved
Optimize test measure by gradient ascent on theoretical E_g (detrimental shift)NTK regression, depth-3 ReLU NTK · MNIST 8s and 9s, M=1000generalization errorHurt
Optimize training distribution over MNIST digits at P=30 rather than uniformNTK regression · MNIST 8s vs 9s, uniform test distributiongeneralization errorImproved
Use a training distribution optimized for small P at larger Poptimized training measure · MNIST 8s vs 9sgeneralization errorHurt
Reduce training-measure rank M_r from 60 to 40 (target power concentrated in 40 directions)linear-kernel regression · Synthetic linear target, N=60, D=120generalization errorMixed
Reduce M_r below 40 (below the target's main directions)linear-kernel regression · Synthetic linear target, N=60, D=120generalization errorHurt
Increase effective regularization (via training variance and M_r)linear regression with label noise · Synthetic, eps^2=0.1double-descent peak and learning speedMixed
Use optimal ridge lambda* = (M_r/D) eps^2linear regression · Synthetic, label noisegeneralization errorImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy