用于黎曼和欧几里得脑电解码的个性化联邦学习
Personalised federated learning for Riemannian and Euclidean EEG decoding
浏览论文内容
中文总结 AI 辅助
本研究提出个性化联邦学习用于EEG解码，让受试者共享主干、保留各自头部，实验表明个性化SPDNet在准确率、收敛速度和通信效率上优于标准FL及集中式训练。
中文摘要 AI 辅助
联邦学习（FL）允许脑电（EEG）解码器从多个受试者的记录中学习，而无需汇集数据。我们考虑两种轻量级EEG解码器：黎曼SPDNet和欧几里得EEGNet。两者都分为主干（trunk）和头部（head），主干用于构建潜在表示，头部用于分类。然而，受试者间的变异性使得单一的共享FL模型难以适配每个受试者。个性化FL解决了这一问题：所有受试者学习一个共同的主干，而每个受试者保留自己的头部。我们将其适配于SPDNet，并研究其相对于标准FL和集中式训练的效果，以EEGNet作为欧几里得基线。实验涵盖三个运动想象数据集，这些数据集在通道、受试者和类别方面覆盖了不同的设置。我们观察到，个性化SPDNet达到了比标准FL和集中式训练更高的准确率，同时收敛轮次更少，且通信参数少于标准FL。在三个数据集中的两个上，它还优于所有EEGNet配置，尽管集中式EEGNet优于集中式SPDNet。
英文摘要
Federated learning (FL) lets EEG decoders learn from recordings of several subjects without pooling them. We consider two light EEG decoders, the Riemannian SPDNet and the Euclidean EEGNet. Both split into a trunk, which builds a latent representation, and a head, which classifies it. Inter-subject variability, however, makes a single shared FL model a poor fit for each subject. Personalised FL addresses this: all subjects learn a common trunk, and each subject keeps its own head. We adapt it for SPDNet and study its effects against standard FL and centralised training, with EEGNet as a Euclidean baseline. Experiments cover three motor-imagery datasets that span diverse regimes in channels, subjects and classes. We observe that personalised SPDNet reaches higher accuracy than both standard FL and centralised training, while converging in fewer rounds and communicating fewer parameters than standard FL. It also outperforms every EEGNet configuration on two of the three datasets, although centralised EEGNet outperforms centralised SPDNet.
发表机构
- CNRS（法国国家科学研究中心）
机构由 AI 辅助整理，请以论文原文为准。
