AI Paper Atlas

Research map / Continual learning / Online and replay-based

Representational Continuity for Unsupervised Continual Learning

ICLR 2022 · Continual learning › Online and replay-based · approach: Unified and unsupervised continual frameworks · 28 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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 unsupervised rather than supervised continual representationscontinual learning strategies · Split CIFAR-10, Split CIFAR-100, Split Tiny-ImageNetforgetting: 71.4\%, 69.7\% and 73.2\% decrease in forgetting with Finetune on Split CIFAR-10, Split CIFAR-100 and Split Tiny-ImageNeHelped
Add LUMP interpolation of current and replay-buffer instancesSimSiam UCL · Split CIFAR-100accuracy: 2.8\% relative increase in accuracyHelped
Add LUMP interpolation of current and replay-buffer instancesSimSiam UCL · Split Tiny-ImageNetaccuracy: 5.9\% relative increase in accuracyHelped
Use unsupervised rather than supervised representationscontinual learning methods · OOD datasets from Split CIFAR-10accuracy: 4.58\%, 6.09\%, 15.26\%, and 17.07\%Helped
Use unsupervised rather than supervised continual learningfew-shot training methods · Split CIFAR-100, 100 training instances per taskaccuracy: 15.78\%p; 32.21\%pHelped

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