Research map / Continual learning / Class-incremental
Forgetting Whenever You Want: A Decentralized Continual Learning Framework with On-Demand Unlearning
Xiao Zhang, Zengzhe Chen, Mingyi Li, Jing Qiao, Fuzhen Zhuang, YUAN YUAN, Dongxiao Yu
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 |
|---|---|---|---|
| add synthetic prototype-guided data D' to local training (ablation) | DCU · RGB Dir(0.5) | accuracy r% after cl0:9: 69.60 -> 75.14 | Improved |
| add replayed synthetic data D-hat for distillation of old classes (ablation) | DCU · RGB Dir(0.5) | accuracy r% after cl10:19: 42.51 -> 59.44 | Improved |
| add synthetic unlearning data D-hat_f guided by prototypes (ablation) | DCU · RGB Dir(0.5) | unlearned-class accuracy u% after ul0: 83.80 -> 0.00 | Improved |
| add prototype-based synthetic data for unlearning historical classes versus FedAF/AU requiring target data | EWC/AU, LwF/AU, FedAF · RGB Dir(0.5) | remaining accuracy r% after ul0: 41.10 -> 58.77 (cl10:19 stage vs DCU ul0) | Improved |
| use gradient ascent on real stored data (QDsga) as sequential tasks arrive | EWC/QDsga, LWF/QDsga · RGB Dir(0.5) | remaining accuracy r% after ul20: LWF/QDsga 5.29 | Hurt |
| use a reservoir-sampling buffer (CLMUL) in decentralized non-IID setting | CLMUL · RGB Dir(0.5) | remaining accuracy r% after ul20: 33.11 | Hurt |
| use diffusion prototypes instead of server-trained generator under stronger non-IID | DCU vs Lander/AU · RGB Dir(0.1) | remaining accuracy r% after ul20: 34.11 -> 40.04 | Improved |
| compare DCU with Lander/AU at mild non-IID | DCU vs Lander/AU · RGB Dir(0.5) | remaining accuracy r% after ul20: 50.67 vs 50.96 | No clear effect |
| standard continual learning methods without unlearning module (EWC, LwF, Target, Lander) | continual learning baselines · RGB and Grayscale | forgetting gap F: e.g. -0.06 to 0.13 | Hurt |
Closest papers by meaning
- Non-exemplar Online Class-Incremental Continual Learning via Dual-Prototype Self-Augment and Refinement (AAAI 2024 · similarity 0.89)
- PDUDT: Provable Decentralized Unlearning under Dynamic Topologies (ICML 2025 · similarity 0.87)
- C$^2$Prompt: Class-aware Client Knowledge Interaction for Federated Continual Learning (NeurIPS 2025 · similarity 0.87)
- Long-Tail Class Incremental Learning via Independent Sub-prototype Construction (CVPR 2024 · similarity 0.86)
- Federated Class-Incremental Learning: A Hybrid Approach Using Latent Exemplars and Data-Free Techniques to Address Local and Global Forgetting (ICLR 2025 · similarity 0.85)
- TinySubNets: An Efficient and Low Capacity Continual Learning Strategy (AAAI 2025 · 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.
