AI Paper Atlas

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

Loss Function Learning for Domain Generalization by Implicit Gradient

Boyan Gao, Henry Gouk, Yongxin Yang, Timothy Hospedales

ICML 2022 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Optimization-based domain generalization · 0 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
replace cross-entropy with learned ITL loss in ERMERM on PACS (ResNet18) · PACSaccuracy (%): +5.5 over ERM+CE (78.9 to 84.4)Improved
replace cross-entropy with learned ITL loss in ERMERM on DomainBed (ResNet50) · DomainBed multi-source average (VLCS, PACS, TerraInc, OfficeHome, DomainNet)accuracy (%): improves on ERM on all listed datasetsImproved
replace cross-entropy with learned ITL loss in ERMERM on single-source DomainBed · Single-source DomainBed (PACS, TerraInc, OfficeHome, VLCS)accuracy (%): PACS 51.85 to 56.54; TerraInc 32.13 to 35.09Improved
replace cross-entropy with ITL loss in RSC (ablation)RSC on PACS (ResNet18) · PACSaccuracy (%): 81.9 to 82.7 (+0.8)Improved
compare ITL to other manually designed robust losses in ERM (ablation)ERM on DomainBed PACS (ResNet50) · DomainBed PACSaccuracy (%): ITL 86.4 vs Focal 84.6, LS 84.9, SCE 84.2, GCE 83.0Improved
learn SCE linear-combination weights with IFT meta-learning (iSCE) instead of Taylor loss (ablation)ERM on PACS (ResNet18) · PACSaccuracy (%): iSCE 82.1 vs SCE 80.3 (improves) but below ITL 84.4Mixed
meta-train loss on RotatedKMNIST instead of RotatedMNIST (ablation)ITL meta-training · OfficeHomeaccuracy (%): 68.74 vs 69.03 (within std)No clear effect
learn loss per task on ColoredMNIST instead of a generic loss (ablation)ITL on ColoredMNIST · ColoredMNISTaccuracy (%): 52.2 vs 52.0 (slight)No clear effect
use implicit gradient (IFT) instead of CMA-ES for meta-training the loss (ablation)ITL meta-training · meta-train loss learningouter loop accuracy: smoother convergence to higher accuracy; K times fewer model trainingsImproved
ITL loss in-domain vs cross-domain PACS testing (ablation)ITL vs CE on PACS · PACS in-domain test accuracyaccuracy (%): 98.96 vs 99.34 in-domain; cross-domain 84.4 vs 81.5Hurt
replace CE with ITL in ERMERM on DomainBed ColoredMNIST · ColoredMNISTaccuracy (%): 52.0 vs 51.5 (+0.5)No 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