AI Paper Atlas

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

Automatic Unsupervised Outlier Model Selection

Yue Zhao, Ryan Rossi, Leman Akoglu

NeurIPS 2021 · Label-efficient and robust learning › Recognition and long-tail · approach: Data augmentation and generalization bias · 26 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
meta-learned model selection (MetaOD) vs always using iForestoutlier detection · POCMAP: 0.1536 -> 0.2035Improved
MetaOD vs always using iForestoutlier detection · STMAP: 0.3197 -> 0.3382Improved
MetaOD selection vs ensemble of all models (ME)outlier detection · POCMAP: 0.1382 -> 0.2035Improved
fix U at meta-feature embedding (MetaOD_F) instead of learning U (ablation)MetaOD · STMAP: 0.3382 -> 0.2594Hurt
concatenate P and meta-features (MetaOD_C) instead of regression approach (ablation)MetaOD · STMAP: 0.3382 -> 0.1946Hurt
MetaOD vs MetaOD_C (ablation)MetaOD · POCMAP: 0.1980 -> 0.2035No clear effect
meta-feature initialization and cyclical learning rate for training stabilityMetaOD · STMAP: 0.2981 -> 0.3382Improved
learned meta-learner (MetaOD) vs global best (GB)model selection · POCMAP: 0.1524 -> 0.2035Improved
local similarity meta-learners vs global best when task similarity is lowAS, ISAC vs GB · STMAP: 0.3137 vs 0.2704Mixed

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