AI Paper Atlas

Research map / Continual learning / Online and replay-based

Helpful or Harmful: Inter-Task Association in Continual Learning

Hyundong Jin, Eunwoo Kim

ECCV 2022 · Continual learning › Online and replay-based · approach: Theoretical analysis of forgetting · 28 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 model search over old tasks (ablation)H2 · Split CIFAR-10 task-incremental T5accuracy (%): 96.9 -> 97.3Improved
add model search over old tasks (ablation)H2 · Split CIFAR-10 class-incremental T5accuracy (%): 79.6 -> 80.3Improved
add Fisher sensitivity measure (ablation)H2 · Split CIFAR-10 task-incremental T5accuracy (%): 96.9 -> 97.3Improved
add Fisher sensitivity measure (ablation)H2 · Split CIFAR-10 class-incremental T5accuracy (%): 78.7 -> 80.3Improved
add search and sensitivity (vs using all old parameters) (ablation)PackNet-style framework · Split CIFAR-10 class-incremental T5accuracy (%): 77.4 -> 80.3Improved
add search and sensitivity (vs using all old parameters) (ablation)PackNet-style framework · Split CIFAR-10 task-incremental T5accuracy (%): 96.8 -> 97.3Improved
add contrastive loss to meta classifier training (ablation)H2 · Split CIFAR-10 class-incremental T5accuracy (%): 77.1 -> 80.3Improved
replace helpful-parameter selection with random selection of old parameters (ablation)H2 · Split CIFAR-10 class-incremental T5accuracy (%): 80.3 -> 68.7Hurt
replace helpful-parameter selection with random selection of old parameters (ablation)H2 · Split CIFAR-10 task-incremental T5accuracy (%): 97.3 -> 96.3Hurt
add search and sensitivity, reducing parameters used per task (ablation)H2 · Split CIFAR-10 parameter consumptionparameters at T5: 4.2x higher than ours for T5 without themImproved
H2 vs PackNet (all old parameters) in task-incrementalH2 · ImageNet + fine-grained avgaccuracy (%): 80.2 -> 82.8Improved
H2 vs TAC in task-incrementalH2 · Split CIFAR-10 task-incrementalaccuracy (%): 96.4 -> 97.3Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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