AI Paper Atlas

Research map / Learning theory and algorithms / Sample complexity

Multi-Fidelity Covariance Estimation in the Log-Euclidean Geometry

Aimee Maurais, Terrence Alsup, Benjamin Peherstorfer, Youssef Marzouk

ICML 2023 · Learning theory and algorithms › Sample complexity · approach: Density and property estimation · 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
Use LEMF instead of high-fidelity-only covariance estimationCovariance estimation · Motivating Gaussian examplelog-Euclidean MSE: 1.72 -> 0.83Improved
Use LEMF instead of truncated Euclidean MFCovariance estimation · Motivating Gaussian examplelog-Euclidean MSE: 66.88 -> 0.83Improved
Use LEMF instead of high-fidelity-only covariance estimationCovariance estimation · Motivating Gaussian exampleaffine-invariant MSE: 4.99 -> 2.69Improved
Use LEMF instead of truncated Euclidean MFCovariance estimation · Motivating Gaussian exampleaffine-invariant MSE: 125.11 -> 2.69Improved
Use LEMF instead of Euclidean MFCovariance estimation · Motivating Gaussian exampleEuclidean MSE: 0.51 -> 0.71Hurt
Combine surrogate samples with high-fidelity samples using LEMFHeat-flow covariance estimation · Heat flowspeedup: 32×Improved
Use LEMF for covariance estimates in metric learningSurface quasi-geostrophy metric learning · Surface quasi-geostrophymean relative error of distances: >20%Improved
Use Euclidean MF rather than LEMFHeat-flow covariance estimation · Heat flowpositive-definiteness validity: more than 10% of the 100 trialsHurt

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