Research map / Continual learning / Online and replay-based
Continual Learning Through Synaptic Intelligence
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 synaptic consolidation (c=1) instead of disabling it (c=0) (ablation) | MLP · Split MNIST | average classification accuracy | Helped |
| Add synaptic consolidation (c=0.1) instead of disabling it (c=0) (ablation) | MLP · Permuted MNIST | average classification accuracy | Helped |
| Use synaptic consolidation rather than ordinary fine-tuning (ablation) | CNN · Split CIFAR-10/CIFAR-100 | validation accuracy | Helped |
| Use synaptic consolidation rather than ordinary fine-tuning (ablation) | CNN · Last task of split CIFAR-10/CIFAR-100 | validation accuracy | Mixed |
| Add synaptic consolidation | MLP · Permuted MNIST | average classification accuracy | Helped |
| Use synaptic consolidation instead of EWC | MLP · Permuted MNIST | average classification accuracy | No clear effect |
| Train on tasks sequentially without synaptic consolidation (ablation) | MLP · Permuted MNIST | average classification accuracy | Hurt |
| Add synaptic consolidation when learning tasks with limited data | CNN · Split CIFAR-10/CIFAR-100 | validation accuracy | Helped |
Closest papers by meaning
- Continual Reinforcement Learning with Complex Synapses (ICML 2018 · similarity 0.93)
- SPICED: A Synaptic Homeostasis-Inspired Framework for Unsupervised Continual EEG Decoding (NeurIPS 2025 · similarity 0.87)
- Learning the Plasticity: Plasticity-Driven Learning Framework in Spiking Neural Networks (NeurIPS 2025 · similarity 0.86)
- Improving Continual Learning Performance and Efficiency with Auxiliary Classifiers (ICML 2025 · similarity 0.86)
- Rapid Learning without Catastrophic Forgetting in the Morris Water Maze (ICML 2024 · similarity 0.86)
- Do Your Best and Get Enough Rest for Continual Learning (CVPR 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.