Irregularly sampled multivariate time series are ubiquitous in several application domains, leading to sparse, not fully-observed and non-aligned observations across different variables. Standard sequential neural network architectures, such as recurrent neural networks (RNNs) and convolutional neural networks (CNNs), consider regular spacing between observation times, posing significant challenges to irregular time series modeling. While most of the proposed architectures incorporate RNN variants to handle irregular time intervals, convolutional neural networks have not been adequately studied in the irregular sampling setting. In this paper, we parameterize convolutional layers by employing time-explicitly initialized kernels. Such general functions of time enhance the learning process of continuous-time hidden dynamics and can be efficiently incorporated into convolutional kernel weights. We, thus, propose the time-parameterized convolutional neural network (TPCNN), which shares similar properties with vanilla convolutions but is carefully designed for irregularly sampled time series. We evaluate TPCNN on both interpolation and classification tasks involving real-world irregularly sampled multivariate time series datasets. Our experimental results indicate the competitive performance of the proposed TPCNN model which is also significantly more efficient than other state-of-the-art methods. At the same time, the proposed architecture allows the interpretability of the input series by leveraging the combination of learnable time functions that improve the network performance in subsequent tasks and expedite the inaugural application of convolutions in this field.
Physiome-ODE: A Benchmark for Irregularly Sampled Multivariate Time Series Forecasting Based on Biological ODEs
ISMLL & VWFS DARC University of Hildesheim（ISMLL与VWFS DARC海德堡大学）；University of Hildesheim（海德堡大学）；Institute of Mathematics TU Berlin（柏林技术大学数学研究所）；DARC University of Hildesheim（DARC海德堡大学）
A Review of Deep Learning Methods for Irregularly Sampled Medical Time Series Data
Key Laboratory of Machine Perception (Ministry of Education), Peking University（北京大学教育部机器感知重点实验室）；School of Electronics Engineering and Computer Science, Peking University（北京大学电子工程与计算机科学学院）；National Institute of Health Data Science, Peking University（北京大学国家健康数据科学研究院）；Institute of Medical Technology, Health Science Center, Peking University（北京大学医学部医学技术研究所）
TE-ESN: Time Encoding Echo State Network for Prediction Based on Irregularly Sampled Time Series Data
Peking University（北京大学）；National Institute of Health Data Science, Peking University（北京大学健康数据科学研究所）；Institute of Medical Technology, Health Science Center of Peking University（北京大学医学部医学技术研究院）
Forecasting West Nile Virus with Graph Neural Networks: Harnessing Spatial Dependence in Irregularly Sampled Geospatial Data
Department of Statistics, University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校统计系）；Department of Statistics, Texas A&M University（德克萨斯A&M大学统计系）；Department of Pathobiology, University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校病理生物学系）
Self-Supervised Transformer for Sparse and Irregularly Sampled Multivariate Clinical Time-Series
Virginia Tech（弗吉尼亚理工大学）
在重症监护环境中经常观察到多变量时间序列数据，其典型特征是稀疏性（缺失信息）和不规则的时间间隔。在这个领域中用于学习表示的现有方法通过聚集或估算值来处理这些挑战，这反过来又抑制了细粒度信息并将不期望的噪声/开销添加到机器学习模型中。为了解决这个问题，我们提出了一个自我监督的Transformer for Time-Series（STraTS）模型，它克服了这些缺陷，将时间序列视为一组观察三元组，而不是使用标准的稠密矩阵表示。它采用了一种新的连续值嵌入技术来编码连续时间和变量值，而无需离散化。它由一个带有多头注意力层的Transformer组件组成，这些层使它能够学习上下文三重嵌入，同时避免循环架构中发生的循环和梯度消失问题。此外，为了解决标记数据可用性有限的问题（这通常在许多医疗保健应用中观察到），STraTS通过利用未标记数据来利用自我监督，以通过使用时间序列预测作为辅助代理任务来学习更好的表示。在真实世界多变量临床时间序列基准数据集上的实验表明，STraTS具有比最先进的死亡率预测方法更好的预测性能，特别是当标记数据有限时。最后，我们还提出了一个可解释的STraTS版本，它可以识别时间序列数据中的重要测量。我们的数据预处理和模型实现代码可以在https://github.com/sindhura97/STraTS上找到。
Graph Neural Flows for Unveiling Systemic Interactions Among Irregularly Sampled Time Series
HES-SO Genève（日内瓦高等教育学院）；University of Manchester（曼彻斯特大学）；Idiap Research Institute（伊迪亚普研究所）；NBC, CRUK Manchester Institute（英国癌症研究中心曼彻斯特研究所（NBC））；MIT-IBM Watson AI Lab（麻省理工学院-IBM沃森人工智能实验室）；IBM Research（IBM研究院）
FlowPath: Learning Data-Driven Manifolds with Invertible Flows for Robust Irregularly-sampled Time Series Classification
Medical & Imaging Informatics (MII) Group, University of California, Los Angeles (UCLA), CA, USA（加州大学洛杉矶分校医学与影像信息学组）；Department of Industrial Engineering, Ulsan National Institute of Science and Technology (UNIST), Republic of Korea（韩国蔚山科学技术院工业工程系）；Artificial Intelligence Graduate School, Ulsan National Institute of Science and Technology (UNIST), Republic of Korea（韩国蔚山科学技术院人工智能研究生院）
ImageFlowNet: Forecasting Multiscale Image-Level Trajectories of Disease Progression with Irregularly-Sampled Longitudinal Medical Images
Yale University（耶鲁大学）；University of California, San Francisco（加州大学旧金山分校）；Mila - Quebec AI Institute（魁北克AI研究所）；Université de Montréal（蒙特利尔大学）；McGill University（麦吉尔大学）；Northern California Retina Vitreous Associates（北加州视网膜与玻璃体协会）
Irregularly Sampled Time Series Interpolation for Binary Evolution Simulations Using Dynamic Time Warping
Electrical and Computer Engineering, Northwestern University（西北大学电气与计算机工程系）；Center for Interdisciplinary Exploration and Research in Astrophysics (CIERA), Northwestern University（跨学科天体物理研究中心（CIERA），西北大学）；NSF-Simons AI Institute for the Sky (SkAI)（国家科学基金会-模拟人工智能天文研究所（SkAI））；Department of Physics and Astronomy, Northwestern University（西北大学物理与天文学系）；Department of Physics, University of Florida（佛罗里达大学物理系）；Institute for Fundamental Theory（基础理论研究所）；Gravitational Wave Science Center (GWSC), Université de Genève（引力波科学中心（GWSC），日内瓦大学）
