arXiv：cs.AI· Abdul Basit, Saim Rehman, Muhammad Shafique·· 13 小时前
ThinkNet：面向高效运动想象 EEG 分类的进化深度集成框架
ThinkNet: Evolutionary Deep Ensembles for Efficient Motor-Imagery EEG Classification
AI 导读
ThinkNet 是一个验证控制的框架，结合仅训练集归一化、验证引导的进化搜索和验证门控推理，用于跨被试运动想象 EEG 解码。在 BCI Competition IV-2a 四分类任务上，其固定基准中验证选出的紧凑解码器仅 4.9K 参数、19 KB FP32 权重，Orin CUDA 单样本推理 0.99 ms，准确率 44.35±15.41%；验证门控集成提升至 43.98±16.25%。
正文
Abstract:Practical assistive and rehabilitative brain--computer interfaces require subject-independent motor-imagery EEG (MI-EEG) decoders that generalize to new users under limited target-user data and constrained compute. However, held-out-subject performance can be overstated when test-subject information influences preprocessing, model selection, or ensemble selection. We present \textit{ThinkNet}, a validation-controlled framework that combines train-only normalization, validation-guided evolutionary search, and validation-gated inference to identify compact decoders and inference policies for held-out subjects. We evaluate four-class BCI Competition IV-2a (session T) decoding with nine Leave-One-Subject-Out (LOSO) folds, three seeds, seven fixed decoder entries, and a broader search over ten representative decoder families; the held-out subject is never used for normalization, hyperparameter, architecture, or ensemble-policy selection. In the fixed benchmark, the validation-selected compact decoder achieved 44.35$\pm$15.41\% accuracy with 4.9K parameters, 19 KB FP32 weights, and 0.99 ms batch-1 Orin CUDA inference. Across the broader search, compact models ($\leq$25K parameters) achieved higher mean held-out accuracy than mid-size and large alternatives after selected retraining (40.10\% vs. 35.09\% and 34.78\%). Validation-gated ensembling improved over validation-selected single-model inference, reaching 43.98$\pm$16.25\% in the fixed benchmark and 43.31$\pm$15.88\% for the compact six-family ensemble. A non-deployable oracle analysis revealed a 6.1-point family-selection gap and near-zero validation--test correlation, showing that validation reliability remains a key bottleneck under subject shift. Thus, ThinkNet is a validation-controlled framework for compact MI-EEG model and inference-policy selection, rather than a single-architecture benchmark.
| Comments: | Accepted to IEEE-EMBS BHI'2026, 7 pages |
| Subjects: | Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Signal Processing (eess.SP) |
| Cite as: | arXiv:2609.33967 [cs.HC] |
| (or arXiv:2609.33967v2 [cs.HC] for this version) | |
| https://doi.org/10.48550/arXiv.2609.33967 arXiv-issued DOI via DataCite |
Submission history
From: Abdul Basit [view email]
[v1]
Sun, 27 Sep 2026 22:11:26 UTC (3,392 KB)
[v2]
Fri, 9 Oct 2026 15:00:12 UTC (3,393 KB)
来源：arXiv：cs.AI · arxiv.org