AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Online Learning from Data Streams with Varying Feature Spaces

Ege Beyazit, Jeevithan Alagurajah, Xindong Wu

AAAI 2019 · Label-efficient and robust learning › Recognition and long-tail · approach: Contrastive representation theory · 62 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 OLVF projection-confidence reweighting instead of OLSFonline linear classifier · IMDB reviewsaverage number of errors: 7851.0 -> 4474.2Improved
use OLVF instead of OLSFonline linear classifier · trapezoidal streams, spambaseaverage number of errors: 993.5 -> 825.8Improved
use OLVF instead of OLSFonline linear classifier · trapezoidal streams, a8aaverage number of errors: 9420.4 -> 8649.8Improved
increase removing ratio of features from 0.25 to 0.75 (ablation)OLVF · spambaseaverage number of errors: 659.8 -> 1375.8Hurt
increase removing ratio of features from 0.25 to 0.75 (ablation)OLVF · wdbcaverage number of errors: 40.8 -> 202.85Hurt
add sparsity projection and truncation to OLVF (ablation)OLVF · spambaseaverage number of errors: 659.8 -> 825.1Hurt
add sparsity projection and truncation to OLVF (ablation)OLVF · wdbcaverage number of errors: 39.6 -> 131Hurt
add sparsity projection and truncation to OLVF (ablation)OLVF · a8aaverage number of errors: 8933.2 -> 8588.6No 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