AI Paper Atlas

Research map / Label-efficient and robust learning / OOD and anomaly detection

ZooD: Exploiting Model Zoo for Out-of-Distribution Generalization

Qishi Dong, Muhammad Awais, Fengwei Zhou, Chuanlong Xie, Tianyang Hu, Yongxin Yang, Sung-Ho Bae, Zhenguo Li

NeurIPS 2022 · Label-efficient and robust learning › OOD and anomaly detection · approach: Out-of-distribution generalization theory · 5 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 the top-1 ranked model with a linear head instead of ERMZooD Single · Office-Homeaccuracy: 66.5 -> 84.6Improved
use the top-1 ranked model instead of SWADZooD Single · PACSaccuracy: 88.1 -> 96.0Improved
use the top-1 ranked model instead of SWADZooD Single · TerraIncognitaaccuracy: 50.0 -> 37.3Hurt
fine-tune the whole top-ranked model instead of only the classifierZooD Single · TerraIncognitaaccuracy: +2.6%Improved
full fine-tuning instead of top-classifier fine-tuning (ablation)ZooD top model · PACSaccuracy: 90.6 -> 96.0Hurt
full fine-tuning instead of top-classifier fine-tuning (ablation)ZooD top model · Office-Homeaccuracy: 83.4 -> 84.6Hurt
ensemble top-K models without feature selection vs single model (ablation)ZooD Ensemble · DomainNetaccuracy: 48.2 -> 50.5Improved
ensemble top-K models without feature selection vs single model (ablation)ZooD Ensemble · TerraIncognitaaccuracy: 37.3 -> 38.2Mixed
add variational EM feature selection to the ensemble (ablation)ZooD · DomainNetaccuracy: 50.5 -> 50.6Improved
add feature selection to the ensemble (ablation)ZooD · TerraIncognitaaccuracy: 38.2 -> 42.3Improved
add feature selection to the ensemble (ablation)ZooD · Office-Homeaccuracy: 85.0 -> 85.1No clear effect
vary number of ensembled models K in {3, 5, 7} (ablation)ZooD · OoD benchmarksaccuracy: K=3No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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