AI Paper Atlas

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

Fast and Accurate Least-Mean-Squares Solvers

Alaa Maalouf, Ibrahim Jubran, Dan Feldman

NeurIPS 2019 · Learning theory and algorithms › Matrix and tensor methods · approach: Low-rank approximation theory · 9 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
run solver on LMS-Coreset (Caratheodory-based compression) instead of full datasklearn Linear/Ridge/Lasso/Elastic-net solvers · synthetic and UCI datasetsrunning time: up to x10; x10--x100 with cross validationImproved
run solver on coreset instead of full dataLMS solvers · synthetic and UCI datasetsaccuracyNo clear effect
use coreset booster instead of SKETCH+CHOLESKY / SKETCH+SVD / SKETCH+INVERSE32-bit float LMS solving · UCI datasets (i),(iii)numerical errorImproved
use coreset booster instead of SKETCH+CHOLESKY32-bit float LMS solving · UCI datasetsrunning time: slower by at most factor of 2Hurt
use faster high-dimension variant (Algorithm with k1,k2, sparsified points) instead of full Caratheodory variant (ablation)LMS-Coreset++ · synthetic and Dataset (iv)running time vs numerical errorImproved
set k2 to (d+1)^2 divided by 12 instead of theoretical optimum (d+1)^2 (ablation)LMS-Coreset++ · synthetic datarunning timeImproved

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