AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Collaborative Optimization and Aggregation for Decentralized Domain Generalization and Adaptation

Guile Wu, Shaogang Gong

ICCV 2021 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Feature alignment for unsupervised adaptation · 91 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
add hybrid batch-instance normalization (ablation)COPA · PACSaccuracy: 82.6 -> 85.1Improved
add collaboration of frozen classifiers (ablation)COPA · PACSaccuracy: 82.7 -> 85.1Improved
add iterative optimization and aggregation (ablation)COPA · PACSaccuracy: 79.6 -> 85.1Improved
add iteration, HBIN and CoFC together (ablation)COPA · PACSaccuracy: 69.7 -> 85.1Improved
replace FedAvg parameter averaging with feature-extractor aggregation plus classifier ensembledecentralized backbone · PACSaccuracy: 77.9 -> 85.1Improved
replace independent training plus logit ensemble with COPAInd-Ensemble · PACSaccuracy: 70.6 -> 85.1Improved
replace centralized joint training with decentralized COPABackbone · PACSaccuracy: 79.5 -> 85.1Improved
replace centralized joint training with decentralized COPABackbone · Digits-DGaccuracy: 73.7 -> 81.5Improved
use RandAugment inside CoFC (ablation)COPA · PACSaccuracyImproved
replace hybrid BN-IN with instance normalization only (ablation)COPA · PACSaccuracyHurt
replace HBIN with IBN-a/IBN-b (ablation)COPA · PACSaccuracyHurt
replace HBIN with Switchable Norm (ablation)COPA · PACSaccuracyNo 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