Research map / Continual learning / Online and replay-based
Task-Free Continual Learning via Online Discrepancy Distance Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 discrepancy-based sample selection to ODDL (ODDL-S) (ablation) | ODDL · Split MNIST | accuracy: 94.85 -> 95.75 | Helped |
| add discrepancy-based sample selection to ODDL (ODDL-S) (ablation) | ODDL · Split CIFAR10 | accuracy: 51.48 -> 52.69 | Helped |
| add discrepancy-based sample selection to ODDL (ODDL-S) (ablation) | ODDL · Split CIFAR100 | accuracy: 26.20 -> 27.21 | Helped |
| add discrepancy-based sample selection to ODDL (ODDL-S) (ablation) | ODDL · Split MImageNet | accuracy: 27.45 -> 28.68 | Helped |
| add discrepancy-based sample selection to ODDL (ODDL-S) (ablation) | ODDL · Permuted MNIST | accuracy: 82.33 -> 83.56 | Helped |
| discrepancy-based expansion criterion for mixture model (ablation) | ODDL-S · Split MNIST | expansion timing | Helped |
| ODDL-S vs strongest dynamic expansion baseline CNDPM | mixture continual learner · Split CIFAR100 | accuracy: 20.10 -> 27.21 | Helped |
| ODDL-S vs random memory removal single model | single classifier · Split MNIST | target risk | Helped |
Closest papers by meaning
- Online Task-Free Continual Learning via Dynamic Expansionable Memory Distribution (CVPR 2025 · similarity 0.90)
- Forgetting, Ignorance or Myopia: Revisiting Key Challenges in Online Continual Learning (NeurIPS 2024 · similarity 0.88)
- Mitigating Catastrophic Forgetting in Online Continual Learning by Modeling Previous Task Interrelations via Pareto Optimization (ICML 2024 · similarity 0.88)
- A Unified and General Framework for Continual Learning (ICLR 2024 · similarity 0.88)
- Do Your Best and Get Enough Rest for Continual Learning (CVPR 2025 · similarity 0.88)
- Doubly Perturbed Task Free Continual Learning (AAAI 2024 · similarity 0.87)
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.