AI Paper Atlas

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

Information Competing Process for Learning Diversified Representations

Jie Hu, Rongrong Ji, ShengChuan Zhang, Xiaoshuai Sun, Qixiang Ye, Chia-Wen Lin, Qi Tian

NeurIPS 2019 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Invariance-based self-supervised learning · 8 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 ICP's two-part diversified representationVGG16 · CIFAR-10 testclassification error (%): 6.67 -> 6.10Improved
use ICP's two-part diversified representationGoogLeNet · CIFAR-10 testclassification error (%): 4.92 -> 4.26Improved
use ICP's two-part diversified representationResNet20 · CIFAR-100 testclassification error (%): 31.91 -> 28.13Improved
use ICP's two-part diversified representationDenseNet40 · CIFAR-100 testclassification error (%): 27.55 -> 24.52Improved
combine information minimization and maximization constraints with competitionβ-VAE · dSpritesMIG: 0.22 -> 0.48Improved
combine information minimization and maximization constraints with competitionβ-VAE · 3D FacesMIG: 0.54 -> 0.73Improved
add ICP reconstruction approachβ-VAE · CelebAMSE: 9.2*10^{-3} -> 8.5*10^{-3}Improved
add ICP reconstruction approachβ-VAE · CelebASSIM: 0.60 -> 0.62Improved
remove information-diversifying and competing constraints (ablation)ICP · dSpritesMIG: 0.48 -> 0.33Hurt
remove information-diversifying and competing constraints (ablation)ICP · 3D FacesMIG: 0.73 -> 0.26Hurt
remove the information-competing objective (ablation)ICP · dSpritesMIG: 0.48 -> 0.20Hurt
remove the information-competing objective (ablation)ICP · 3D FacesMIG: 0.73 -> 0.57Hurt

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