Statistical Deep Learning for Spatial and Spatiotemporal Data

被引:20
|
作者
Wikle, Christopher K. [1 ]
Zammit-Mangion, Andrew [2 ]
机构
[1] Univ Missouri, Dept Stat, Columbia, MO 65211 USA
[2] Univ Wollongong, Sch Math & Appl Stat, Wollongong, NSW, Australia
基金
澳大利亚研究理事会; 美国国家科学基金会;
关键词
Bayesian hierarchical models; convolutional neural networks; deep Gaussian processes; recurrent neural networks; reinforcement learning; warping; CONVOLUTIONAL NEURAL-NETWORKS; DATA ASSIMILATION; NONSTATIONARY; EMULATOR; MODELS; INFERENCE;
D O I
10.1146/annurev-statistics-033021-112628
中图分类号
O1 [数学];
学科分类号
0701 ; 070101 ;
摘要
Deep neural network models have become ubiquitous in recent years and have been applied to nearly all areas of science, engineering, and industry. These models are particularly useful for data that have strong dependencies in space (e.g., images) and time (e.g., sequences). Indeed, deep models have also been extensively used by the statistical community to model spatial and spatiotemporal data through, for example, the use of multilevel Bayesian hierarchical models and deep Gaussian processes. In this review, we first present an overview of traditional statistical and machine learning perspectives for modeling spatial and spatiotemporal data, and then focus on a variety of hybrid models that have recently been developed for latent process, data, and parameter specifications. These hybrid models integrate statistical modeling ideas with deep neural network models in order to take advantage of the strengths of each modeling paradigm. We conclude by giving an overview of computational technologies that have proven useful for these hybrid models, and with a brief discussion on future research directions.
引用
收藏
页码:247 / 270
页数:24
相关论文
共 50 条
  • [41] Learning to learn from data: Using deep adversarial learning to construct optimal statistical procedures
    Luedtke, Alex
    Carone, Marco
    Simon, Noah
    Sofrygin, Oleg
    SCIENCE ADVANCES, 2020, 6 (09):
  • [42] The impacts of training data spatial resolution on deep learning in remote sensing
    Ardohain, Christopher
    Fei, Songlin
    SCIENCE OF REMOTE SENSING, 2025, 11
  • [43] Applying statistical learning theory to deep learning
    Gerbelot, Cedric
    Karagulyan, Avetik
    Karp, Stefani
    Ravichandran, Kavya
    Stern, Menachem
    Srebro, Nathan
    JOURNAL OF STATISTICAL MECHANICS-THEORY AND EXPERIMENT, 2024, 2024 (10):
  • [44] Statistical analysis of spatial point patterns on deep seismic reflection data: a preliminary test
    Vasudevan, K.
    Eckel, S.
    Fleischer, F.
    Schmidt, V.
    Cook, F. A.
    GEOPHYSICAL JOURNAL INTERNATIONAL, 2007, 171 (02) : 823 - 840
  • [45] A Statistical Learning Model with Deep Learning Characteristics
    Liao, Lei
    Huang, Zhiqiu
    Wang, Wengjie
    51ST ANNUAL IEEE/IFIP INTERNATIONAL CONFERENCE ON DEPENDABLE SYSTEMS AND NETWORKS (DSN-W 2021), 2021, : 137 - 140
  • [46] Statistical learning of spatiotemporal target regularities in the absence of saliency
    Xu, Zhenzhen
    Theeuwes, Jan
    Los, Sander A.
    ATTENTION PERCEPTION & PSYCHOPHYSICS, 2025, 87 (02) : 431 - 444
  • [47] Spatiotemporal Deep Learning for Bridge Response Forecasting
    Zhang, Ruiyang
    Meng, Libo
    Mao, Zhu
    Sun, Hao
    JOURNAL OF STRUCTURAL ENGINEERING, 2021, 147 (06)
  • [48] Learning deep spatiotemporal features for video captioning
    Daskalakis, Eleftherios
    Tzelepi, Maria
    Tefas, Anastasios
    PATTERN RECOGNITION LETTERS, 2018, 116 : 143 - 149
  • [49] Analytics of deep model-based spatiotemporal and spatial feature learning methods for surgical action classification
    Rachana S. Oza
    Mayuri A. Mehta
    Ketan Kotecha
    Jerry Chun-Wei Lin
    Multimedia Tools and Applications, 2024, 83 : 52275 - 52303
  • [50] Spatiotemporal prediction of carbon emissions using a hybrid deep learning model considering temporal and spatial correlations
    Chen, Yixiang
    Xie, Yuxin
    Dang, Xu
    Huang, Bo
    Wu, Chao
    Jiao, Donglai
    ENVIRONMENTAL MODELLING & SOFTWARE, 2024, 172