AI Paper Atlas

Research map / Continual learning / Class-incremental

Semantic Drift Compensation for Class-Incremental Learning

Lu Yu, Bartlomiej Twardowski, Xialei Liu, Luis Herranz, Kai Wang, Yongmei Cheng, Shangling Jui, Joost van de Weijer

CVPR 2020 · Continual learning › Class-incremental · approach: Class-balanced knowledge retention · 320 citations (OpenAlex)

Check 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 embedding-network fine-tuning with metric learning instead of softmax classificationClass-incremental learning · CUB-200-2011average incremental accuracy: +23.9%Improved
Use embedding-network fine-tuning with metric learning instead of softmax classificationClass-incremental learning · CIFAR100average incremental accuracy: +32.6%Improved
Use embedding-network fine-tuning with metric learning instead of softmax classificationClass-incremental learning · ImageNet-Subsetaverage incremental accuracy: +42.9%Improved
Add SDC to E-FT (ablation)E-FT · CIFAR100, T11average incremental accuracy: 37.4 -> 44.8Improved
Add SDC to E-FT (ablation)E-FT · ImageNet-Subset, T11average incremental accuracy: 47.4 -> 50.9Improved
Add SDC to E-LwF (ablation)E-LwF · CIFAR100, T11average incremental accuracy: 46.1 -> 46.1No clear effect
Add SDC to E-LwF (ablation)E-LwF · ImageNet-Subset, T11average incremental accuracy: 51.5 -> 50.5Hurt
Add SDC to E-EWC (ablation)E-EWC · CIFAR100, T11average incremental accuracy: 40.8 -> 46.1Improved
Add SDC to E-EWC (ablation)E-EWC · ImageNet-Subset, T11average incremental accuracy: 49.5 -> 51.5Improved
Add SDC to E-MAS (ablation)E-MAS · CIFAR100, T11average incremental accuracy: 43.1 -> 46.3Improved
Add SDC to E-MAS (ablation)E-MAS · ImageNet-Subset, T11average incremental accuracy: 50.8 -> 51.2Improved
Replace triplet loss with Multi-similarity loss (ablation)E-FT · CUB-200-2011, T6average incremental accuracy: +13.3%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