AI Paper Atlas

Research map / Learning theory and algorithms / Sample complexity

Optimal Estimator for Unlabeled Linear Regression

Hang Zhang, Ping Li

ICML 2020 · Learning theory and algorithms › Sample complexity · approach: Coresets and sublinear estimation

Read the paper

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
add alternating minimization refinement (iterative updates of Π and B) after the one-step estimate (ablation)single-observation unlabeled regression (m=1, p=2) estimator · n=1000, p=2, noiseless syntheticpermutation error ||Π̂-Π*||_2: from about 13000 to below 2000Improved
increase the number of observations m (stable rank of B*) from small to large valuesproposed one-step estimator · simulations n=500 and n=1000required SNR for recovery successImproved
the estimator is applied with a single observation (m=1) and p>1 without known direction of B*proposed one-step estimator · m=1, p>1exact permutation recoveryHurt
larger number of permuted rows h or larger dimension p (fixed n)proposed one-step estimator · simulations n=500, n=1000recovery difficulty (required SNR)Hurt

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