AI Paper Atlas

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

Memory Efficient Kernel Approximation

Si Si, Cho-Jui Hsieh, Inderjit S. Dhillon

JMLR 2017 · Learning theory and algorithms › Matrix and tensor methods · approach: Structured low-rank learning

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 k-means clustering in input space instead of random partition for blocks (ablation)BKA / MEKA blocks · ijcnn1 subsample (4000 points, 5 blocks)numerical rank of blocks: k-means block ranks 7-29 versus random partition ranks 43-139Improved
use k-means partition instead of random partition in BKA (ablation)BKA · covtype and cadata (500 samples, five partitions)Gaussian kernel approximation errorImproved
add off-diagonal link matrix (MEKA) instead of block-diagonal only BKA (ablation)BKA · covtype sample, varying gammarelative approximation errorImproved
increase number of clusters in BKA-SVM (ablation)BKA-SVM · ijcnn1, census, covtypetraining time and accuracyMixed
use BKA-SVM instead of LLSVM and FastFood low-rank solverskernel SVM · covtypetest accuracy (%): 84.21 -> 96.05 (LLSVM), 80.1 -> 96.05 (FastFood)Improved
use MEKA instead of standard Nystrom at equal memoryNystrom · ijcnn1 gamma=1relative kernel approximation error: 0.0423 -> 0.0082Improved
use MEKA instead of standard Nystrom at equal memoryNystrom · covtype gamma=10relative kernel approximation error: 0.3700 -> 0.1192Improved
use Nystrom-based basis instead of random feature (RKS, Fastfood) methodskernel approximation · pendigit, ijcnn1, covtyperelative kernel approximation error: 0.0811 vs 0.4404 (RKS), 0.4726 (fastfood) on pendigitImproved
use MEKA instead of standard Nystrom for kernel ridge regressionNystrom · covtype kernel ridge regressiontest RMSE: 0.8197 -> 0.7106Improved
use MEKA instead of standard Nystrom for kernel ridge regressionNystrom · cpusmall kernel ridge regressiontest RMSE: 8.8747 -> 6.1130Improved
raise thresholding epsilon that zeros off-diagonal link blocks (ablation)MEKA · ijcnn1 and cadatatime vs approximation errorMixed
choose rank per cluster by size or by eigenvalues instead of the same k (ablation)MEKA · ijcnn1 and cadataapproximation errorNo clear effect

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