Papers › Helpful or Harmful: Inter-Task Association in Continual Learning
Helpful or Harmful: Inter-Task Association in Continual Learning
Hyundong Jin, Eunwoo Kim
When optimizing sequentially incoming tasks, deep neural networks generally suffer from catastrophic forgetting due to their lack of ability to maintain knowledge from old tasks. This may lead to a significant performance drop of the previously learned tasks. To alleviate this problem, studies on continual learning have been conducted as a countermeasure. Nevertheless, it suffers from an increase in computational cost due to the expansion of the network size or a change in knowledge that is favorably linked to previous tasks. In this work, we propose a novel approach to differentiate helpful and harmful information for old tasks using a model search to learn a current task effectively. Given a new task, the proposed method discovers an underlying association knowledge from old tasks, which can provide additional support in acquiring the new task knowledge. In addition, by introducing a sensitivity measure to the loss of the current task from the associated tasks, we find cooperative relations between tasks while alleviating harmful interference. We apply the proposed approach to both task- and class-incremental scenarios in continual learning, using a wide range of datasets from small to large scales. Experimental results show that the proposed method outperforms a large variety of continual learning approaches for the experiments while effectively alleviating catastrophic forgetting.
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Continual Learning | CUBS (Fine-grained 6 Tasks) | H$^{2}$ | Accuracy | 84.1 | #2 of 6 | Archive leaderboard | report |
| Continual Learning | Flowers (Fine-grained 6 Tasks) | H$^{2}$ | Accuracy | 94.9 | #3 of 6 | Archive leaderboard | report |
| Continual Learning | ImageNet (Fine-grained 6 Tasks) | H$^{2}$ | Accuracy | 75.71 | #6 of 6 | Archive leaderboard | report |
| Continual Learning | Sketch (Fine-grained 6 Tasks) | H$^{2}$ | Accuracy | 76.2 | #5 of 6 | Archive leaderboard | report |
| Continual Learning | Split CIFAR-10 (5 tasks) | H$^{2}$ | Top 1 Accuracy % | 97.3 | #1 of 1 | Archive leaderboard | report |
| Continual Learning | Split MNIST (5 tasks) | H$^{2}$ | Top 1 Accuracy % | 99.9 | #1 of 1 | Archive leaderboard | report |
| Continual Learning | Stanford Cars (Fine-grained 6 Tasks) | H$^{2}$ | Accuracy | 90.6 | #3 of 6 | Archive leaderboard | report |
| Continual Learning | Wikiart (Fine-grained 6 Tasks) | H$^{2}$ | Accuracy | 75.1 | #3 of 6 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections