Research map / Continual learning / Online and replay-based
Forget-free Continual Learning with Winning Subnetworks
Haeyong Kang, Rusty John Lloyd Mina, Sultan Rizky Hikmawan Madjid, Jaehong Yoon, Mark Hasegawa-Johnson, Sung Ju Hwang, Chang D. Yoo
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use WSN with task-adaptive subnetworks and selectively reused weights | continual learning · Permuted MNIST | ACC (%): 96.37~ ($\pm$ 0.1) -> 96.41~ ($\pm$ 0.1) | Improved |
| Use WSN with task-adaptive subnetworks and selectively reused weights | continual learning · Omniglot Rotation | ACC (%): 30.70~ ($\pm$ 1.5) -> 87.28}~ ($\pm$ 0.7) | Improved |
| Use WSN with task-adaptive subnetworks and selectively reused weights | continual learning · TinyImageNet | ACC (%): 67.39~ ($\pm$ 0.5) -> 71.96}~ ($\pm$ \textbf{1.4}) | Improved |
| Increase selected WSN capacity from c=3% to c=30% (ablation) | WSN · Permuted MNIST | ACC (%): 94.84 -> 96.41 | Improved |
| Increase selected WSN capacity from c=3% to c=50% (ablation) | WSN · 5 Datasets | ACC (%): 90.57 -> 93.41 | Improved |
| Use SoftNet soft masks instead of binary WSN masks (ablation) | continual learning · CIFAR-100, session 9 | accuracy: 46.31 -> 46.63 | Improved |
| Use SoftNet soft masks instead of binary WSN masks (ablation) | continual learning · miniImageNet, session 9 | accuracy: 45.66 -> 50.48 | Improved |
Closest papers by meaning
- Continual Learning: Forget-Free Winning Subnetworks for Video Representations (TPAMI 2024 · similarity 0.91)
- On the Soft-Subnetwork for Few-Shot Class Incremental Learning (ICLR 2023 · similarity 0.87)
- Enhancing Knowledge Transfer for Task Incremental Learning with Data-free Subnetwork (NeurIPS 2023 · similarity 0.87)
- TinySubNets: An Efficient and Low Capacity Continual Learning Strategy (AAAI 2025 · similarity 0.86)
- Task-aware Orthogonal Sparse Network for Exploring Shared Knowledge in Continual Learning (ICML 2024 · similarity 0.85)
- Layerwise Proximal Replay: A Proximal Point Method for Online Continual Learning (ICML 2024 · similarity 0.85)
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.
