arXiv cs.AI· Muyun Jiang, Yi Ding, Wei Zhang, Jinbo Chen, Chenyu Liu, Zhenjie Yang, Yuxin Li, Jingyuan Chen, Yuhao Lu, Yong Li, Shuailei Zhang, Cuntai Guan·· 4 小时前AI 评分35
AutoBCI：面向 EEG 脑机接口的预测引导式智能体神经架构发现框架
AutoBCI: Forecast-Guided Agentic Neural Architecture Discovery for EEG-Based Brain-Computer Interfaces
AI 导读
研究者提出 AutoBCI，一个由 Designer Agent 和 Forecaster Agent 协同完成 EEG 解码架构发现与选择的智能体框架。用 Claude Opus 5.5 等 6 个 LLM 在 14 个 EEG 数据集上评测，发现的架构平均测试 balanced accuracy 达 64.16%，优于最强基线 REVE 的 63.87%。
正文 · 原文
Authors:Muyun Jiang, Yi Ding, Wei Zhang, Jinbo Chen, Chenyu Liu, Zhenjie Yang, Yuxin Li, Jingyuan Chen, Yuhao Lu, Yong Li, Shuailei Zhang, Cuntai Guan
Abstract:EEG-based brain-computer interfaces support a broad range of applications, yet designing decoding architectures that perform well across diverse tasks remains challenging. We introduce AutoBCI, an agentic framework in which a Designer Agent and a Forecaster Agent support the discovery and selection of EEG decoding architectures across tasks. The Designer Agent performs Pool-Guided Architecture Discovery (PGAD), generating and refining architectures through training and validation across multiple EEG tasks, such as emotion recognition, motor imagery, and sleep staging. The Forecaster Agent performs Performance Estimation from Early Knowledge (PEEK), using architecture code, the training protocol, and early learning curves to predict full-budget validation performance and select promising candidates for continued training. Across 14 EEG datasets spanning motor imagery, emotion recognition, and sleep staging, we evaluate AutoBCI with six LLMs, including Opus 5.5 and GPT 5.6 Sol, and compare the architectures selected by the search procedure against ten baselines: six conventional EEG models and four foundation models. The architecture discovered by AutoBCI with Claude Opus 5.5 achieves 64.16% average test balanced accuracy (bAcc), compared with 63.87% for REVE, the strongest baseline on this metric. Using ten observed epochs, PEEK reduces mean absolute error in predicting average validation bAcc from 2.20 to 1.36 percentage points, a 38.1% reduction relative to the best-observed-score baseline. Code is available at this https URL
| Comments: | 34 pages, 6 figures, including supplementary material |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2609.35456 [cs.AI] |
| (or arXiv:2609.35456v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2609.35456 arXiv-issued DOI via DataCite |
Submission history
From: Muyun Jiang [view email]
[v1]
Mon, 28 Sep 2026 15:46:12 UTC (1,669 KB)
[v2]
Wed, 4 Nov 2026 02:48:15 UTC (1,669 KB)
来源：arXiv cs.AI · arxiv.org