AI Paper Atlas

Research map / Continual learning / Online and replay-based

IIRC: Incremental Implicitly-Refined Classification

Mohamed Abdelsalam, Mojtaba Faramarzi, Shagun Sodhani, Sarath Chandar

CVPR 2021 · Continual learning › Online and replay-based · approach: Dynamic expansion and isolation · 23 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 iCaRL-CNN/iCaRL-norm instead of other baselineslifelong learners · IIRC-CIFAR/ImageNetpw-JSImproved
use cosine normalization (iCaRL-norm) instead of iCaRL-CNNiCaRL · IIRC-ImageNetpw-JSImproved
learn the scalar eta instead of fixing it (ablation)iCaRL-norm · IIRCaccuracyImproved
use A-GEM instead of vanilla ERER · IIRC-CIFARpw-JSHurt
keep all old samples but with incomplete information (ER-infinite) instead of complete information (incremental joint)replay · IIRCpw-JSHurt
measure with pw-JS instead of JSiCaRL family · IIRC-CIFARmetric scoreHurt
train on more subsequent tasksER, LUCIR baselines · IIRCper-task performanceHurt
use LUCIR instead of iCaRL-norm for superclass associationclass incremental · IIRC-CIFARsuperclass labellingMixed
use ER with buffer of 100 samples per class (ablation)ER · IIRC-ImageNetaverage performanceNo clear effect
vary replay buffer size per class (ablation)ER · IIRC-CIFAR/ImageNetaverage performanceMixed

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