AI Paper Atlas

Research map / Continual learning / Online and replay-based

Continual Learning of a Mixed Sequence of Similar and Dissimilar Tasks

Zixuan Ke, Bing Liu, Xingchang Huang

NeurIPS 2020 · Continual learning › Online and replay-based · approach: Theoretical analysis of forgetting · 53 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
Use CAT with task masks, similarity detection, and KTAcontinual learning of mixed task sequences · Four mixed-sequence datasets, 2-layer fully connected networkoverall accuracy: 0.7293 vs 0.7337; 0.6194 vs 0.5967; 0.8439 vs 0.8245; 0.6843 vs 0.6796Improved
Use CAT with task masks, similarity detection, and KTAcontinual learning of mixed task sequences · M(CIFAR100-10, F-CelebA) and M(CIFAR100-20, F-CelebA), AlexNet-like architectureoverall accuracy: 0.6764 -> 0.6831; 0.7428 -> 0.7468Improved
Add backward knowledge transfer to forward transfer (ablation)CAT · F-EMNIST and F-CelebA tasks in four mixed sequencesaccuracy: 0.6104 -> 0.6134; 0.6081 -> 0.6187; 0.6873 -> 0.6909; 0.6782 -> 0.7164Improved
Add KTA (ablation)CAT · Overall accuracy across four mixed-sequence datasetsaccuracy: 0.7585 -> 0.7710; 0.5915 -> 0.6194; 0.8423 -> 0.8439; 0.6572 -> 0.6843Improved
Detect task similarity instead of treating all previous tasks as dissimilar (ablation)CAT · Overall accuracy across four mixed-sequence datasetsaccuracy: 0.7337 -> 0.7710; 0.5967 -> 0.6194; 0.8245 -> 0.8439; 0.6796 -> 0.6843Improved
Detect task similarity instead of treating all previous tasks as similar (ablation)CAT · Overall accuracy across four mixed-sequence datasetsaccuracy: 0.1887 -> 0.7710; 0.3090 -> 0.6194; 0.3631 -> 0.8439; 0.3207 -> 0.6843Improved
Add task masks and similarity detection to the all-sim, no-KTA variant (ablation)CAT · Overall accuracy across four mixed-sequence datasetsaccuracy: 0.7293 -> 0.7710; 0.5535 -> 0.6194; 0.8024 -> 0.8439; 0.6018 -> 0.6843Improved
Add KTA (ablation)CAT · Dissimilar-task subsets in mixed sequencesaccuracy: 0.9301 -> 0.9287; 0.5594 -> 0.5479; 0.9637 -> 0.9566; 0.6740 -> 0.6683Mixed
Transfer selectively from similar prior tasks using KTA (ablation)CAT · Similar-task subsets in four mixed sequencesaccuracy: 0.5870 -> 0.6134; 0.6236 -> 0.6909; 0.5995 -> 0.6187; 0.6236 -> 0.7164Improved

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