In recent years, Transformer-based auto-attention mechanisms have been successfully applied to the analysis of a variety of context-reliant data types, from texts to images and beyond, including data from non-Euclidean geometries. In this paper, we present such a mechanism, designed to classify sequences of Symmetric Positive Definite matrices while preserving their Riemannian geometry throughout the analysis. We apply our method to automatic sleep staging on timeseries of EEG-derived covariance matrices from a standard dataset, obtaining high levels of stage-wise performance.
Fusion of Physiological and Behavioural Signals on SPD Manifolds with Application to Stress and Pain Detection
CNRS(法国国家科学研究中心);University of Lille(里尔大学);Centrale Lille(里尔中央理工学院);IMT Nord Europe(法国北部高等矿业电信学院);Institut Mines-Télécom(法国矿业电信学院);INSERM(法国国家健康与医学研究院)
Learning to Normalize on the SPD Manifold under Bures-Wasserstein Geometry
School of Artificial Intelligence and Computer Science, Jiangnan University(江南大学人工智能与计算机学院);Department of Information Engineering and Computer Science, University of Trento(特伦托大学信息工程与计算机科学系)
Riemannian Networks over Full-Rank Correlation Matrices
Department of Information Engineering and Computer Science, University of Trento, Trento, Italy(特伦托大学信息工程与计算机科学系);School of Artificial Intelligence and Computer Science, Jiangnan University, Wuxi, China(江南大学人工智能与计算机科学学院)
Deep Optimal Transport for Domain Adaptation on SPD Manifolds
College of Computing and Data Science, Nanyang Technological University(计算与数据科学学院,南洋理工大学);Université Paris-Saclay, Inria, CEA, Palaiseau, France(巴黎-萨克雷大学,法国国家信息与自动化技术研究所,法国原子能委员会,Palaiseau)
Learning Log-Determinant Divergences for Positive Definite Matrices
Mitsubishi Electric Research Labs (MERL)(三菱电机研究实验室);Australian Centre for Robotic Vision, Australian National University(澳大利亚国立大学澳大利亚机器人视觉中心);University of Minnesota(明尼苏达大学);Research School of Engineering, The Australian National University(澳大利亚国立大学工程研究学院);Monash University(莫纳什大学);Data61-CSIRO(澳大利亚联邦科学与工业研究组织数据61)
Towards Robust EEG Decoding Based on Riemannian Self-Attention
School of Artificial Intelligence and Computer Science(人工智能与计算机科学学院);Jiangnan University(江南大学);Department of Information Engineering and Computer Science(信息工程与计算机科学系);University of Trento(特伦托大学);Centre for Vision, Speech and Signal Processing (CVSSP)(视觉、语音与信号处理中心);University of Surrey(萨里大学)
A Unified SPD Token Transformer Framework for EEG Classification: Systematic Comparison of Geometric Embeddings
Independent Researcher(独立研究者);Department of Electrophysics, National Yang Ming Chiao Tung University, Hsinchu, Taiwan(国家阳明交通大学电子物理系);Department of Mathematics, Boise State University, Boise, ID, USA(数学系,博伊州立大学)
Wrapped Gaussian on the manifold of Symmetric Positive Definite Matrices
LAMSADE, CNRS, PSL Univ. Paris-Dauphine, France(巴黎-萨克雷大学巴黎-达菲学院、国家科学研究中心、巴黎-达菲大学);Inria center at the University of Bordeaux / LaBRI, France(波尔多大学/LaBRI与Inria中心);TAU, LISN, University Paris-Saclay, France(巴黎-萨克雷大学、LISN、TAU)