AI Paper Atlas

Research map / Continual learning / Class-incremental

Online Prototype Learning for Online Continual Learning

Yujie Wei, Jiaxin Ye, Zhizhong Huang, Junping Zhang, Hongming Shan

ICCV 2023 · Continual learning › Class-incremental · approach: Class-balanced knowledge retention · 77 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 OnPro (online prototype equilibrium + adaptive prototypical feedback) over the best baseline OCMonline class-incremental learning · CIFAR-10 (M=0.1k)average accuracy: 47.5 -> 57.8Improved
add online prototype equilibrium (OPE) to the baseline (ablation)OnPro · CIFAR-10 (M=0.1k)average accuracy: 52.0 -> 57.8Improved
add adaptive prototypical feedback (APF) (ablation)OnPro · CIFAR-100 (M=0.5k)average accuracy: 19.3 -> 22.7Improved
replace random mixup with APF-guided sampling (ablation)OnPro · CIFAR-10 (M=0.1k)average accuracy: 53.5 -> 57.8Improved
include both new-class and seen-class prototype losses in OPE (ablation)OnPro · CIFAR-10 (M=0.1k)average accuracy: 54.8 / 55.7 -> 57.8Improved
let seen-class prototype loss include new classes (ablation)OnPro · CIFAR-100 (M=0.5k)average accuracy: 20.8 -> 22.7Improved
add knowledge distillation to ERER · CIFAR-10 (M=0.1k)average accuracy: 19.4 -> 17.0Hurt
add projection head to OnPro (ablation)OnPro · CIFAR-10 (M=0.1k)average accuracy: 56.1 -> 57.8Improved
increase APF ratio alpha from 0 to 0.25 (ablation)OnPro · CIFAR-10 (M=0.2k)average accuracy: 62.9 -> 65.5Improved
increase APF ratio alpha beyond 0.1 (ablation)OnPro · CIFAR-100 (M=0.5k)average accuracy: 22.7 -> 21.1Hurt
add rotation augmentation to baselinesER, ASER, DVC · CIFAR-10average accuracy: SCR-Rot 35.8 vs SCR 40.2 at M=0.1k; ER-Rot 30.1 vs ER 19.4Mixed
OnPro vs baselines on average forgetting on TinyImageNetOnPro · TinyImageNetaverage forgetting: OnPro 21.3 vs iCaRL 9.9 and CoPE 11.9 at M=1kNo 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