Attention to Warp: Deep Metric Learning for Multivariate Time Series

被引:1
|
作者
Matsuo, Shinnosuke [1 ]
Wu, Xiaomeng [2 ]
Atarsaikhan, Gantugs [1 ]
Kimura, Akisato [2 ]
Kashino, Kunio [2 ]
Iwana, Brian Kenji [1 ]
Uchida, Seiichi [1 ]
机构
[1] Kyushu Univ, Fukuoka, Japan
[2] NTT Corp, Commun Sci Labs, Tokyo, Japan
关键词
Attention model; Dynamic time warping; Metric learning; Signature verification; ONLINE SIGNATURE VERIFICATION; ALGORITHM;
D O I
10.1007/978-3-030-86334-0_23
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Deep time series metric learning is challenging due to the difficult trade-off between temporal invariance to nonlinear distortion and discriminative power in identifying non-matching sequences. This paper proposes a novel neural network-based approach for robust yet discriminative time series classification and verification. This approach adapts a parameterized attention model to time warping for greater and more adaptive temporal invariance. It is robust against not only local but also large global distortions, so that even matching pairs that do not satisfy the monotonicity, continuity, and boundary conditions can still be successfully identified. Learning of this model is further guided by dynamic time warping to impose temporal constraints for stabilized training and higher discriminative power. It can learn to augment the inter-class variation through warping, so that similar but different classes can be effectively distinguished. We experimentally demonstrate the superiority of the proposed approach over previous non-parametric and deep models by combining it with a deep online signature verification framework, after confirming its promising behavior in single-letter handwriting classification on the Unipen dataset.
引用
收藏
页码:350 / 365
页数:16
相关论文
共 50 条
  • [1] Time-frequency deep metric learning for multivariate time series classification
    Chen, Zhi
    Liu, Yongguo
    Zhu, Jiajing
    Zhang, Yun
    Jin, Rongjiang
    He, Xia
    Tao, Jing
    Chen, Lidian
    [J]. NEUROCOMPUTING, 2021, 462 : 221 - 237
  • [2] Multivariate Time Series Early Classification with Interpretability Using Deep Learning and Attention Mechanism
    Hsu, En-Yu
    Liu, Chien-Liang
    Tseng, Vincent S.
    [J]. ADVANCES IN KNOWLEDGE DISCOVERY AND DATA MINING, PAKDD 2019, PT III, 2019, 11441 : 541 - 553
  • [3] Benchmarking Attention-Based Interpretability of Deep Learning in Multivariate Time Series Predictions
    Baric, Domjan
    Fumic, Petar
    Horvatic, Davor
    Lipic, Tomislav
    [J]. ENTROPY, 2021, 23 (02) : 1 - 23
  • [4] Anomaly Detection of Multivariate Time Series Based on Metric Learning
    Wang, Hongkai
    Feng, Jun
    Peng, Liangying
    Pan, Sichen
    Zhao, Shuai
    Jin, Helin
    [J]. DATA SCIENCE (ICPCSEE 2022), PT I, 2022, 1628 : 94 - 110
  • [5] Deep Multiple Metric Learning for Time Series Classification
    Chen, Zhi
    Liu, Yongguo
    Zhu, Jiajing
    Zhang, Yun
    Li, Qiaoqin
    Jin, Rongjiang
    He, Xia
    [J]. IEEE ACCESS, 2021, 9 : 17829 - 17842
  • [6] Multivariate Financial Time Series Forecasting with Deep Learning
    Martelo, Sebastian
    Leon, Diego
    Hernandez, German
    [J]. APPLIED COMPUTER SCIENCES IN ENGINEERING, WEA 2022, 2022, 1685 : 160 - 169
  • [7] Order-Preserving Metric Learning for Mining Multivariate Time Series
    Xu, Jie
    Xu, Zhenxing
    Yu, Bin
    Wang, Fei
    [J]. 20TH IEEE INTERNATIONAL CONFERENCE ON DATA MINING (ICDM 2020), 2020, : 711 - 720
  • [8] Adversarial Examples in Deep Learning for Multivariate Time Series Regression
    Mode, Gautam Raj
    Hoque, Khaza Anuarul
    [J]. 2020 IEEE APPLIED IMAGERY PATTERN RECOGNITION WORKSHOP (AIPR): TRUSTED COMPUTING, PRIVACY, AND SECURING MULTIMEDIA, 2020,
  • [9] Deep attention fuzzy cognitive maps for interpretable multivariate time series prediction
    Qin, Dunwang
    Peng, Zhen
    Wu, Lifeng
    [J]. KNOWLEDGE-BASED SYSTEMS, 2023, 275
  • [10] A new metric for classification of multivariate time series
    Guan, Heshan
    Jiang, Qingshan
    Hong, Zhiling
    [J]. FOURTH INTERNATIONAL CONFERENCE ON FUZZY SYSTEMS AND KNOWLEDGE DISCOVERY, VOL 1, PROCEEDINGS, 2007, : 453 - +