提出三种高效方法，解决脑电听觉注意力解码的初始化偏差问题
Efficient Solutions for Mitigating Initialization Bias in Unsupervised Self-Adaptive Auditory Attention Decoding
- 设计无需标签的自适应解码框架，避免用户校准
- 在多个数据集上实现与现有方法相当的准确率（92.1%~94.7%）
- 计算开销恒定，适合实时神经听力设备部署
从多说话人环境下的脑电图（EEG）中解码关注的说话人近年来受到广泛关注，其驱动应用为神经引导的助听设备。当前方法通常依赖训练时的关注说话人真实标签，需为每位用户和每套EEG设备进行校准。尽管已开发出无需标签的自适应听觉注意力解码（AAD）用于刺激重建，但存在初始化偏差，影响性能。虽有无偏变体被提出，但其计算复杂度随数据量增加而上升。本文提出三种计算高效的替代方案，在保持相当性能的同时，计算成本显著降低且恒定。实验在MAESTRO、ECoG-ASR等数据集上验证，准确率可达92.1%~94.7%。代码已公开于https://github.com/YYao-42/Unsupervised_AAD。
原文摘要 · Abstract (English)
Decoding the attended speaker in a multi-speaker environment from electroencephalography (EEG) has attracted growing interest in recent years, with neuro-steered hearing devices as a driver application. Current approaches typically rely on ground-truth labels of the attended speaker during training, necessitating calibration sessions for each user and each EEG set-up to achieve optimal performance. While unsupervised self-adaptive auditory attention decoding (AAD) for stimulus reconstruction has been developed to eliminate the need for labeled data, it suffers from an initialization bias that can compromise performance. Although an unbiased variant has been proposed to address this limitation, it introduces substantial computational complexity that scales with data size. This paper presents three computationally efficient alternatives that achieve comparable performance, but with a significantly lower and constant computational cost. The code for the proposed algorithms is available at https://github.com/YYao-42/Unsupervised_AAD.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。