Research map / Continual learning / Class-incremental
From Selection to Scheduling: Federated Geometry-Aware Correction Makes Exemplar Replay Work Better under Continual Dynamic Heterogeneity
Zhuang Qi, Ying-Peng Tang, Lei Meng, Guoqing Chao, Lei Wu, Han Yu, Xiangxu Meng
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 fixed ETF classifier alone (ablation) | FedCBDR · CIFAR10 3 tasks beta=0.5, K=5 | Top-1 accuracy (%): 63.32 -> 62.17 | Hurt |
| add fixed ETF classifier alone (ablation) | FedCBDR · CIFAR100 5 tasks beta=0.1, K=5 | Top-1 accuracy (%): 45.84 -> 44.58 | Hurt |
| add GSA on top of ETF (ablation) | FedCBDR+ETF · CIFAR10 3 tasks beta=0.5, K=5 | Top-1 accuracy (%): 62.17 -> 68.54 | Improved |
| add EGC on top of ETF (ablation) | FedCBDR+ETF · CIFAR10 3 tasks beta=0.5, K=5 | Top-1 accuracy (%): 62.17 -> 69.12 | Improved |
| add EGC on top of ETF+GSA (ablation) | FedCBDR+ETF+GSA · CIFAR10 3 tasks beta=0.5, K=5 | Top-1 accuracy (%): 68.54 -> 72.67 | Improved |
| add GSA on top of ETF+EGC (ablation) | FedCBDR+ETF+EGC · CIFAR100 5 tasks beta=0.1, K=5 | Top-1 accuracy (%): 47.16 -> 50.14 | Improved |
| plug FEAT into FedCBDR (FEAT_F) | FedCBDR · CIFAR10 3 tasks beta=0.5, K=5 | Top-1 accuracy (%): 63.32 -> 72.67 | Improved |
| plug FEAT into Re-Fed+ (FEAT_R) | Re-Fed+ · CIFAR100 5 tasks beta=0.1, K=5 | Top-1 accuracy (%): 31.92 -> 37.14 | Improved |
| add FedRCL to FedCBDR | FedCBDR · CIFAR10 5 tasks beta=0.5, K=5 | Top-1 accuracy (%): 61.77 -> 58.74 | Hurt |
| add MOON to FedCBDR | FedCBDR · CIFAR100 5 tasks beta=0.1, K=5 | Top-1 accuracy (%): 45.84 -> 45.14 | No clear effect |
| add FEAT to FedCBDR instead of FedFSA | FedCBDR · CIFAR10 5 tasks beta=0.5, K=5 | Top-1 accuracy (%): 64.42 -> 70.19 | Improved |
| add GSA and EGC to FedCBDR, last task accuracy on CIFAR10 | FedCBDR · CIFAR10 3 tasks beta=0.5, Task 3 accuracy | Top-1 accuracy (%): 96.80 -> 94.28 | Hurt |
Closest papers by meaning
- Class-wise Balancing Data Replay for Federated Class-Incremental Learning (NeurIPS 2025 · similarity 0.88)
- Accurate Forgetting for Heterogeneous Federated Continual Learning (ICLR 2024 · similarity 0.88)
- Towards Efficient Replay in Federated Incremental Learning (CVPR 2024 · similarity 0.87)
- Federated Class-Incremental Learning: A Hybrid Approach Using Latent Exemplars and Data-Free Techniques to Address Local and Global Forgetting (ICLR 2025 · similarity 0.87)
- Federated Continual Learning Goes Online: Uncertainty-Aware Memory Management for Vision Tasks and Beyond (ICLR 2025 · similarity 0.87)
- CAN: Leveraging Clients As Navigators for Generative Replay in Federated Continual Learning (ICML 2025 · similarity 0.86)
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.
