AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Learning Representation for Multitask Learning through Self-Supervised Auxiliary Learning

Seokwon Shin, Hyungrok Do, Youngdoo Son

ECCV 2024 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Invariance-based self-supervised learning · 3 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 dummy gradient norm regularization to Vanilla MTLVanilla MTL (ResNet50, Split) · UTKFaceDelta_MTL: +2.66% -> +3.50%Improved
add DGR to CAGradCAGrad · UTKFaceDelta_MTL: +3.17% -> +4.11%Improved
add DGR to PCGradPCGrad · UTKFaceDelta_MTL: +2.80% -> +2.50%Hurt
add DGR to Vanilla MTLVanilla MTL (ResNet-50, Split) · NYUv2Delta_MTL: +6.32% -> +8.26%Improved
add DGR to PCGradPCGrad (Split) · NYUv2Delta_MTL: +4.70% -> +7.19%Improved
add DGR to Uncertainty weighting with MTANUncertainty (MTAN) · NYUv2Delta_MTL: +7.57% -> +7.28%Hurt
add DGR to IMTLIMTL · NYUv2Delta_MTL: +13.75% -> +14.13% (Split)Improved
add DGR to DWA with MTANDWA (MTAN) · CityScapesDelta_MTL: +4.43% -> +4.31%Hurt
use 3 dummy decoders instead of 1 or 5 (ablation)DGR · NYUv2 (3 tasks)Delta_MTL: d=1 +7.13%, d=3 +8.05%, d=5 +7.96%Improved
apply DGR to single-task ImageNet with 5 dummy decoders (ablation)ResNet-50 · ImageNettop-1 accuracy: 76.13% -> 75.98%Hurt
add DGR with a Swin backboneVanilla MTL (Swin) · NYUv2Delta_MTL: +6.74% -> +8.77%Improved
add DGR on Pascal with 5 tasksVanilla MTL · PascalDelta_MTL: -2.86% -> -2.42%Improved

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