Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.15877 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910223695872000 |
|---|---|
| author | Vahedifar, Mohammad Ali Ray, Abhisek Zhang, Qi |
| author_facet | Vahedifar, Mohammad Ali Ray, Abhisek Zhang, Qi |
| contents | Continual learning enables neural networks to learn tasks sequentially without forgetting previously acquired knowledge. However, neural networks suffer from catastrophic forgetting, where learning new tasks degrades performance on earlier ones. We address this problem with Shapley Neuron Valuation (SNV), a principled framework that quantifies Neuron importance in continual learning, grounded in cooperative game theory. SNV selectively freezes important Neurons while keeping others plastic, enabling buffer-free continual learning without expanding architecture. Experiments on ImageNet-1k show that SNV consistently outperforms existing buffer-free methods. In particular, SNV improves accuracy by +2.88% in the class incremental learning and +6.46% in the task incremental learning scenarios compared to the second baseline. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_15877 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Shapley Neuron Values for Continual Learning: Which Neurons Matter Most? Vahedifar, Mohammad Ali Ray, Abhisek Zhang, Qi Machine Learning Artificial Intelligence Continual learning enables neural networks to learn tasks sequentially without forgetting previously acquired knowledge. However, neural networks suffer from catastrophic forgetting, where learning new tasks degrades performance on earlier ones. We address this problem with Shapley Neuron Valuation (SNV), a principled framework that quantifies Neuron importance in continual learning, grounded in cooperative game theory. SNV selectively freezes important Neurons while keeping others plastic, enabling buffer-free continual learning without expanding architecture. Experiments on ImageNet-1k show that SNV consistently outperforms existing buffer-free methods. In particular, SNV improves accuracy by +2.88% in the class incremental learning and +6.46% in the task incremental learning scenarios compared to the second baseline. |
| title | Shapley Neuron Values for Continual Learning: Which Neurons Matter Most? |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2605.15877 |