Research map / Continual learning / Online and replay-based
Adaptive Plasticity Improvement for Continual 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 |
|---|---|---|---|
| Adaptively expand layer inputs using GRR while retaining DualGPM gradient projection | API · CIFAR100-sup | ACC (%): 58.2 ± 0.2 -> 60.2 ± 0.2 | Helped |
| Adaptively expand layer inputs using GRR while retaining DualGPM gradient projection | API · Split CIFAR100 | ACC (%): 80.5 ± 0.3 -> 81.4 ± 0.4 | Helped |
| Adaptively expand layer inputs using GRR while retaining DualGPM gradient projection | API · Split Mini-Imagenet | ACC (%): 62.5 ± 0.7 -> 65.9 ± 0.6 | Helped |
| Adaptively expand layer inputs using GRR while retaining DualGPM gradient projection | API · 5-Datasets | ACC (%): 90.9 ± 0.1 -> 91.1 ± 0.3 | Helped |
| Add adaptive plasticity improvement to DualGPM (ablation) | DualGPM · CIFAR100-sup | ACC (%): 57.6 ± 0.7 -> 60.2 ± 0.2 | Helped |
| Add adaptive plasticity improvement to DualGPM (ablation) | DualGPM · Split CIFAR100 | ACC (%): 78.5 ± 0.4 -> 81.4 ± 0.4 | Helped |
| Add adaptive plasticity improvement to DualGPM (ablation) | DualGPM · Split Mini-Imagenet | ACC (%): 61.2 ± 0.6 -> 65.9 ± 0.6 | Helped |
| Add adaptive plasticity improvement to DualGPM (ablation) | DualGPM · 5-Datasets | ACC (%): 88.7 ± 0.5 -> 91.1 ± 0.3 | Helped |
| Replace DualGPM with GPM (ablation) | API · Split CIFAR100 | MEM (M): 2.0 -> 7.3 | Hurt |
| Replace DualGPM with GPM (ablation) | API · 5-Datasets | MEM (M): 3.1 -> 7.7 | Hurt |
| Use GRR-guided expansion instead of equal channel expansion (ablation) | API · Split CIFAR100 | Param (M): 0.60 -> 0.26 | Helped |
| Use GRR-guided expansion instead of equal channel expansion (ablation) | API · 5-Datasets | Param (M): 0.19 -> 0.11 | Helped |
Closest papers by meaning
- Improving Plasticity in Online Continual Learning via Collaborative Learning (CVPR 2024 · similarity 0.92)
- A Unified and General Framework for Continual Learning (ICLR 2024 · similarity 0.90)
- Pareto Continual Learning: Preference-Conditioned Learning and Adaption for Dynamic Stability-Plasticity Trade-off (AAAI 2025 · similarity 0.89)
- Re-evaluating Continual Learning with Few-Shot Adaptation (NeurIPS 2026 · similarity 0.89)
- BECAME: Bayesian Continual Learning with Adaptive Model Merging (ICML 2025 · similarity 0.89)
- STAR: Stability-Inducing Weight Perturbation for Continual Learning (ICLR 2025 · similarity 0.89)
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.