AI Paper Atlas

Research map / Learning theory and algorithms / Matrix and tensor methods

Regression with Label Permutation in Generalized Linear Model

Guanhua Fang, Ping Li

ICML 2023 · Learning theory and algorithms › Matrix and tensor methods · approach: High-dimensional estimation theory · 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
increase number of observations m per individual toward about 10 log2 nML estimation for label permutation in GLM (Poisson-type simulations) · simulation, n=256/512probability of successful permutation recovery: transition from 0 to 1 near m ≈ 10 log2 nImproved
decrease observation rate q (more missing entries)ML and two-step estimation with missing observations · simulationprobability of successful permutation recoveryHurt
increase proportion of permuted labels beyond about 20%two-step estimation · simulation Setting 2probability of successful permutation recovery: drops from 1 to 0 around 20% wrong labelsHurt
fit linear model to raw or log-transformed count-type data instead of GLM (ablation)permutation recovery under model misspecification · simulation Setting 3recovery performanceHurt
increase number of responses m for theory of known-B case with Gaussian design (Example 2 and 3)permutation recovery when B is known · theory, complete/sparse design examplesrequired m: m ≳ log n suffices vs m=1 impossibleImproved

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