Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.17439 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912598915547136 |
|---|---|
| author | Zhou, Yangxuan Zhao, Sha Wang, Jiquan Jiang, Haiteng Li, Shijian Li, Tao Pan, Gang |
| author_facet | Zhou, Yangxuan Zhao, Sha Wang, Jiquan Jiang, Haiteng Li, Shijian Li, Tao Pan, Gang |
| contents | Human brain achieves dynamic stability-plasticity balance through synaptic homeostasis. Inspired by this biological principle, we propose SPICED: a neuromorphic framework that integrates the synaptic homeostasis mechanism for unsupervised continual EEG decoding, particularly addressing practical scenarios where new individuals with inter-individual variability emerge continually. SPICED comprises a novel synaptic network that enables dynamic expansion during continual adaptation through three bio-inspired neural mechanisms: (1) critical memory reactivation; (2) synaptic consolidation and (3) synaptic renormalization. The interplay within synaptic homeostasis dynamically strengthens task-discriminative memory traces and weakens detrimental memories. By integrating these mechanisms with continual learning system, SPICED preferentially replays task-discriminative memory traces that exhibit strong associations with newly emerging individuals, thereby achieving robust adaptations. Meanwhile, SPICED effectively mitigates catastrophic forgetting by suppressing the replay prioritization of detrimental memories during long-term continual learning. Validated on three EEG datasets, SPICED show its effectiveness. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_17439 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SPICED: A Synaptic Homeostasis-Inspired Framework for Unsupervised Continual EEG Decoding Zhou, Yangxuan Zhao, Sha Wang, Jiquan Jiang, Haiteng Li, Shijian Li, Tao Pan, Gang Artificial Intelligence Machine Learning Human brain achieves dynamic stability-plasticity balance through synaptic homeostasis. Inspired by this biological principle, we propose SPICED: a neuromorphic framework that integrates the synaptic homeostasis mechanism for unsupervised continual EEG decoding, particularly addressing practical scenarios where new individuals with inter-individual variability emerge continually. SPICED comprises a novel synaptic network that enables dynamic expansion during continual adaptation through three bio-inspired neural mechanisms: (1) critical memory reactivation; (2) synaptic consolidation and (3) synaptic renormalization. The interplay within synaptic homeostasis dynamically strengthens task-discriminative memory traces and weakens detrimental memories. By integrating these mechanisms with continual learning system, SPICED preferentially replays task-discriminative memory traces that exhibit strong associations with newly emerging individuals, thereby achieving robust adaptations. Meanwhile, SPICED effectively mitigates catastrophic forgetting by suppressing the replay prioritization of detrimental memories during long-term continual learning. Validated on three EEG datasets, SPICED show its effectiveness. |
| title | SPICED: A Synaptic Homeostasis-Inspired Framework for Unsupervised Continual EEG Decoding |
| topic | Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2509.17439 |