AI Paper Atlas

Research map / Optimization / Optimizers

Large Scale Online Kernel Learning

Jing Lu, Steven C.H. Hoi, Jialei Wang, Peilin Zhao, Zhi-Yong Liu

JMLR 2016 · Optimization › Optimizers · approach: Surrogate losses and robust solvers · 140 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 Nystrom approximation (NOGD) instead of random-removal budget (BSGD-R)budget SGD · codrna batchaccuracy (%): 67.16 -> 95.92Improved
use NOGD instead of BSGD-M mergingbudget SGD · codrna batchtrain time (s): 132.50 -> 2.55Improved
use NOGD instead of exact LIBSVMLIBSVM · KDD99 batchtrain time (s): 20772.10 -> 20.71Improved
use FOGD instead of kernel OGDOGD · w7a online binarymistake rate (%): 2.96 -> 2.75Improved
use FOGD instead of BOGDBOGD · a9a online binarymistake rate (%): 27.9 -> 17.4Improved
use FOGD instead of kernel OGDOGD · a9a online binarytime (s): 1549.5 -> 1.8Improved
use NOGD instead of BOGDBOGD · acoustic multi-classmistake rate (%): 58.0 -> 37.8Improved
use NOGD instead of FOGD at small budget B=100FOGD · german online binarymistake rate (%): 29.9 -> 30.4No clear effect
use NOGD instead of FOGD at large budget in batchFOGD · codrna batchaccuracy (%): 94.24 -> 95.92Improved
increase rho_f or rho_n (more Fourier components or rank) (ablation)FOGD and NOGD · multi-class datasets (B=200)mistake rateMixed
use Projectron++ instead of BOGD (projection vs removal)BOGD · usps multi-classmistake rate (%): 23.2 -> 9.9Improved
use BPA-S in multi-class instead of binaryBPA-S · multi-classtime costHurt

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