AI Paper Atlas

Research map / Continual learning / Class-incremental

A Theoretical Study on Solving Continual Learning

Gyuhak Kim, Changnan Xiao, Tatsuya Konishi, Zixuan Ke, Bing Liu

NeurIPS 2022 · Continual learning › Class-incremental · approach: Class-balanced knowledge retention · 16 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
replace ODIN with CSI as the per-task OOD detector (ablation)HAT · C100-10TCIL accuracy: 41.2 -> 63.3Improved
replace ODIN with CSI as the per-task OOD detector (ablation)Sup · C100-20TCIL accuracy: 36.4 -> 60.2Improved
replace ODIN with CSI as the per-task OOD detector (ablation)HAT · C10-5TCIL accuracy: 62.6 -> 87.8Improved
apply ODIN post-processingDER++ · C100-10TCIL accuracy: 53.71 -> 55.29Improved
apply ODIN post-processingSup · C100-10TCIL accuracy: 44.58 -> 46.74Improved
apply ODIN post-processingBiC · C100-10TCIL accuracy: 52.92 -> 48.65Hurt
apply ODIN post-processingPASS · C100-10TCIL accuracy: 33.00 -> 31.00Hurt
apply ODIN post-processingLwF · C100-10TCIL accuracy: 45.26 -> 51.82 CIL while AUC 88.30 -> 87.11Mixed
add output calibration with small memory buffer (ablation)HAT+CSI · C100-20TCIL accuracy: 54.6 -> 58.0Improved
add output calibration with small memory buffer (ablation)Sup+CSI · M-5TCIL accuracy: 80.7 -> 81.0No clear effect
add CSI-based OOD detection to HATHAT · C100-10T TIL accuracyTIL accuracy: 84.0 -> 92.0Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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