A journal of IEEE and CAA , publishes high-quality papers in English on original theoretical/experimental research and development in all areas of automation
Advanced Search
Volume 13 Issue 10
Oct.  2026

IEEE/CAA Journal of Automatica Sinica

  • JCR Impact Factor: 18.3, Top 1 (SCI Q1)
    CiteScore: 28.2, Top 1% (Q1)
    Google Scholar h5-index: 95， TOP 5
Turn off MathJax
Article Contents
Article Navigation >  IEEE/CAA Journal of Automatica Sinica  > 2026  >  13(10): 2280-2299
H. Wang, L. Pan, Y. Lu, Z. Li, S. He, X. Jiang, Z. Chen, and X. Liu, “FMT-I: Flexible mass transport for noise-resilient missing data imputation,” IEEE/CAA J. Autom. Sinica, vol. 13, no. 10, pp. 2280–2299, Oct. 2026. doi: 10.1109/JAS.2026.125876
Citation: H. Wang, L. Pan, Y. Lu, Z. Li, S. He, X. Jiang, Z. Chen, and X. Liu, “FMT-I: Flexible mass transport for noise-resilient missing data imputation,” IEEE/CAA J. Autom. Sinica, vol. 13, no. 10, pp. 2280–2299, Oct. 2026. doi: 10.1109/JAS.2026.125876
shu

FMT-I: Flexible Mass Transport for Noise-Resilient Missing Data Imputation

doi: 10.1109/JAS.2026.125876
Funds:  This work was supported by the National Natural Science Foundation of China (62606018) and the China Postdoctoral Science Foundation (2025M781449)
More Information
    Abstract
  • The vulnerability of existing missing data imputation methods to noisy observations often limits their utility in real-world industrial applications. To address this, we introduce a noise-resilient approach to missing data imputation based on optimal transport theory. Specifically, we first present the flexible mass transport (FMT) discrepancy, which quantifies the discrepancy between two distributions as the minimal cost to match samples between them. Unlike standard transport problems, FMT allows for matching a flexible mass of each sample, thereby aligning reliable samples while excluding noisy ones. Subsequently, we propose the FMT for imputation (FMT-I) framework, which iteratively minimizes the FMT discrepancy between arbitrary mini-batch samples and utilizes the resulting gradients to update imputed values. This process ensures the consistency of statistical properties between the imputed data and the entire dataset, thereby producing feasible imputations. Experiments on real-world industrial datasets confirm that FMT-I exhibits strong noise resilience and outperforms state-of-the-art imputation methods, offering a reliable solution for missing data imputation problems in complex industrial contexts.

     

    • FullText(HTML)
  • loading
    • References(70)
  • [1]
    H. Wang, Z. Wang, Y. Niu, Z. Liu, H. Li, Y. Liao, Y. Huang, and X. Liu, “An accurate and interpretable framework for trustworthy process monitoring,” IEEE Trans. Artif. Intell., vol. 5, no. 5, pp. 2241−2252, May 2024. doi: 10.1109/TAI.2023.3319606
    [2]
    Z. Chen, H. Wang, Z. Song, and Z. Ge, “Improving data-driven inferential sensor modeling by industrial knowledge: A Bayesian perspective,” IEEE Trans. Syst. Man Cybern.: Syst., vol. 55, no. 2, pp. 1043−1055, Feb. 2025. doi: 10.1109/TSMC.2024.3493071
    [3]
    L. Pan, H. Wang, Z. Chen, Y. Huang, Y. Niu, Z. Liu, Q. He, and X. Liu, “Controllable mixture-of-experts for multivariate soft sensors,” IEEE Trans. Autom. Sci. Eng., vol. 22, pp. 19789−19800, Aug. 2025. doi: 10.1109/TASE.2025.3597838
    [4]
    Z. Chen, Y. Zhang, O. Zhang, F. Wang, L. Yao, H. Wang, and Z. Song, “Blending data and knowledge for process industrial modeling under Riemannian preconditioned Bayesian framework,” IEEE Trans. Knowl. Data Eng., vol. 38, no. 1, pp. 82−95, Jan. 2026. doi: 10.1109/TKDE.2025.3621125
    [5]
    Z. Zhang, S. Ye, Y. Zhang, W. Ding, and H. Wang, “Belief combination of classifiers for incomplete data,” IEEE/CAA J. Autom. Sinica, vol. 9, no. 4, pp. 652−667, Apr. 2022. doi: 10.1109/JAS.2022.105458
    [6]
    Z. Chen, H. Wang, G. Chen, Y. Ma, L. Yao, Z. Ge, and Z. Song, “Analyzing and improving supervised nonlinear dynamical probabilistic latent variable model for inferential sensors,” IEEE Trans. Ind. Informat., vol. 20, no. 11, pp. 13296−13307, Nov. 2024. doi: 10.1109/TII.2024.3435466
    [7]
    H. Wang, Z. Chen, Z. Liu, L. Pan, H. Xu, Y. Liao, H. Li, and X. Liu, “SPOT-I: Similarity preserved optimal transport for industrial IoT data imputation,” IEEE Trans. Ind. Informat., vol. 20, no. 12, pp. 14421−14429, Dec. 2024. doi: 10.1109/TII.2024.3452241
    [8]
    H. Wang, X. Liu, Z. Liu, H. Li, Y. Liao, Y. Huang, and Z. Chen, “LSPT-D: Local similarity preserved transport for direct industrial data imputation,” IEEE Trans. Autom. Sci. Eng., vol. 22, pp. 9438−9448, 2025. doi: 10.1109/TASE.2024.3506835
    [9]
    D. Jarrett, B. Cebere, T. Liu, A. Curth, and M. van der Schaar, “HyperImpute: Generalized iterative imputation with automatic model selection,” in Proc. Int. Conf. Machine Learning, Baltimore, USA, 2022, pp. 9916−9937.
    [10]
    Z. Zhang, “Missing data imputation: Focusing on single imputation,” Ann. Transl. Med., vol. 4, no. 1, Art. no. 9, Jan. 2016.
    [11]
    P. Royston and I. R. White, “Multiple imputation by chained equations (MICE): Implementation in Stata,” J. Statist. Softw., vol. 45, no. 4, pp. 1−20, Dec. 2011. doi: 10.18637/jss.v045.i04
    [12]
    Y. Ouyang, L. Xie, C. Li, and G. Cheng, “MissDiff: Training diffusion models on tabular data with missing values,” arXiv preprint arXiv: 2307.00467, 2023.
    [13]
    Y. Tashiro, J. Song, Y. Song, and S. Ermon, “CSDI: Conditional score-based diffusion models for probabilistic time series imputation,” in Proc. 35th Int. Conf. Neural Information Processing Systems, 2021, Art. no. 1900.
    [14]
    J. Yoon, J. Jordon, and M. van der Schaar, “GAIN: Missing data imputation using generative adversarial nets,” in Proc. 35th Int. Conf. Machine Learning, Stockholm, Sweden, 2018, pp. 5675−5684.
    [15]
    T. Wang, H. Ke, A. Jolfaei, S. Wen, M. S. Haghighi, and S. Huang, “Missing value filling based on the collaboration of cloud and edge in artificial intelligence of things,” IEEE Trans. Ind. Informat., vol. 18, no. 8, pp. 5394−5402, Aug. 2022. doi: 10.1109/TII.2021.3126110
    [16]
    H. Li, C. Zheng, W. Wang, H. Wang, F. Feng, and X.-H. Zhou, “Debiased recommendation with noisy feedback,” in Proc. 30th ACM SIGKDD Conf. Knowledge Discovery and Data Mining, Barcelona, Spain, 2024, pp. 1576−1586.
    [17]
    H. Wang, Z. Chen, H. Wang, Y. Tan, P. Li, T. Liu, X. Chen, H. Li, and Z. Lin, “Unbiased recommender learning from implicit feedback via weakly supervised learning,” in Proc. 42nd Int. Conf. Machine Learning, Vancouver, Canada, 2025.
    [18]
    D. J. Stekhoven and P. Bühlmann, “MissForest—non-parametric missing value imputation for mixed-type data,” Bioinformatics, vol. 28, no. 1, pp. 112−118, Jan. 2012. doi: 10.1093/bioinformatics/btr597
    [19]
    H. Chen, Y. Jiang, S. Guo, X. Mao, Y. Lin, and H. Wan, “DiffLight: A partial rewards conditioned diffusion model for traffic signal control with missing data,” in Proc. 38th Int. Conf. Neural Information Processing Systems, Vancouver, Canada, 2024, Art. no. 3921.
    [20]
    Y. Zhang, X. Kong, W. Zhou, J. Liu, Y. Fu, and G. Shen, “A comprehensive survey on traffic missing data imputation,” IEEE Trans. Intell. Transp. Syst., vol. 25, no. 12, pp. 19252−19275, Dec. 2024. doi: 10.1109/TITS.2024.3478816
    [21]
    H. Li, Y. Liao, Z. Tian, Z. Liu, J. Liu, and X. Liu, “Bidirectional stackable recurrent generative adversarial imputation network for specific emitter missing data imputation,” IEEE Trans. Inf. Forensics Security, vol. 19, pp. 2967−2980, Jan. 2024. doi: 10.1109/TIFS.2024.3352393
    [22]
    Y. Du, C. Yang, N. Yu, W. Lin, Q. Zhao, and S. Wang, “Latent imputation before prediction: A new computational paradigm for de novo peptide sequencing,” in Proc. 42nd Int. Conf. Machine Learning, Vancouver, Canada, 2025.
    [23]
    D. Um, J. W. Yoon, S. J. Ahn, and Y. Yeo, “Gene-gene relationship modeling based on genetic evidence for single-cell RNA-seq data imputation,” in Proc. 38th Int. Conf. Neural Information Processing Systems, Vancouver, Canada, 2024, Art. no. 598.
    [24]
    R. Malarvizhi and A. S. Thanamani, “K-nearest neighbor in missing data imputation,” Int. J. Eng. Res. Dev., vol. 5, no. 1, pp. 5−7, Nov. 2012.
    [25]
    R. Mazumder, T. Hastie, and R. Tibshirani, “Spectral regularization algorithms for learning large incomplete matrices,” J. Mach. Learn. Res., vol. 11, pp. 2287−2322, Mar. 2010.
    [26]
    P.-A. Mattei and J. Frellsen, “MIWAE: Deep generative modelling and imputation of incomplete data sets,” in Proc. 36th Int. Conf. Machine Learning, Long Beach, USA, 2019, pp. 4413−4423.
    [27]
    T. Kyono, Y. Zhang, A. Bellot, and M. van der Schaar, “MIRACLE: Causally-aware imputation via learning missing data mechanisms,” in Proc. 35th Int. Conf. Neural Information Processing Systems, 2021, Art. no. 1823.
    [28]
    Y. Liu and A. C. Constantinou, “Improving the imputation of missing data with Markov blanket discovery,” in Proc. 11th Int. Conf. Learning Representations, Kigali, Rwanda, 2022.
    [29]
    J. Fan, C. Yang, and M. Udell, “Robust non-linear matrix factorization for dictionary learning, denoising, and clustering,” IEEE Trans. Signal Process., vol. 69, pp. 1755−1770, Mar. 2021. doi: 10.1109/TSP.2021.3062988
    [30]
    Y. Hu and D. B. Work, “Robust tensor recovery with fiber outliers for traffic events,” ACM Trans. Knowl. Discov. Data, vol. 15, no. 1, Art. no. 6, Feb. 2021. doi: 10.1145/3417337
    [31]
    L. Hu, Y. Jia, W. Chen, L. Wen, and Z. Ye, “A flexible and robust tensor completion approach for traffic data recovery with low-rankness,” IEEE Trans. Intell. Transp. Syst., vol. 25, no. 3, pp. 2558−2572, Mar. 2023. doi: 10.1109/tits.2023.3319033
    [32]
    X. Feng, H. Zhang, C. Wang, and H. Zheng, “Traffic data recovery from corrupted and incomplete observations via spatial-temporal TRPCA,” IEEE Trans. Intell. Transp. Syst., vol. 23, no. 10, pp. 17835−17848, Oct. 2022. doi: 10.1109/TITS.2022.3151925
    [33]
    J. Kim, K. Lee, and T. Park, “To predict or not to predict? Proportionally masked autoencoders for tabular data imputation,” in Proc. 39th AAAI Conf. Artificial Intelligence, Washington, USA, 2025, pp. 17886−17894.
    [34]
    T. Du, L. Melis, and T. Wang, “ReMasker: Imputing tabular data with masked autoencoding,” in Proc. 12th Int. Conf. Learning Representations, Vienna, Austria, 2024.
    [35]
    Z. Pan, Y. Wang, K. Wang, H. Chen, C. Yang, and W. Gui, “Imputation of missing values in time series using an adaptive-learned median-filled deep autoencoder,” IEEE Trans. Cybern., vol. 53, no. 2, pp. 695−706, Feb. 2023. doi: 10.1109/TCYB.2022.3167995
    [36]
    M. Kang, R. Zhu, D. Chen, X. Liu, and W. Yu, “CM-GAN: A cross-modal generative adversarial network for imputing completely missing data in digital industry,” IEEE Trans. Neural Netw. Learn. Syst., vol. 35, no. 3, pp. 2917−2926, Mar. 2024. doi: 10.1109/TNNLS.2023.3284666
    [37]
    Z. Sun, H. Li, W. Wang, J. Liu, and X. Liu, “DTIN: Dual transformer-based imputation nets for multivariate time series emitter missing data,” Knowl. Based Syst., vol. 284, Art. no. 111270, Jan. 2024. doi: 10.1016/j.knosys.2023.111270
    [38]
    Y. Luo, X. Cai, Y. Zhang, J. Xu, and X. Yuan, “Multivariate time series imputation with generative adversarial networks,” in Proc. 32nd Int. Conf. Neural Inf. Process. Syst., Red Hook, USA, 2018, pp. 1603−1614.
    [39]
    Z. Chen, H. Li, F. Wang, O. Zhang, H. Xu, X. Jiang, Z. Song, and H. Wang, “Rethinking the diffusion models for missing data imputation: A gradient flow perspective,” in Proc. 38th Int. Conf. Neural Inf. Process. Syst., Vancouver, Canada, 2024. Art. no. 3558.
    [40]
    C. Ou, H. Zhu, Y. A. W. Shardt, L. Ye, X. Yuan, Y. Wang, C. Yang, and W. Gui, “Missing-data imputation with position-encoding denoising autoencoders for industrial processes,” IEEE Trans. Instrum. Meas., vol. 73, Art. no. 2528911, Aug. 2024.
    [41]
    D. Liu, Y. Wang, C. Liu, K. Wang, X. Yuan, and C. Yang, “Blackout missing data recovery in industrial time series based on masked-former hierarchical imputation framework,” IEEE Trans. Autom. Sci. Eng., vol. 21, no. 2, pp. 1138−1150, Apr. 2024. doi: 10.1109/TASE.2023.3287895
    [42]
    X.-Y. Li, Y. Xu, Q.-X. Zhu, and Y.-L. He, “Industrial data imputation based on multiscale spatiotemporal information embedding with asymmetrical transformer,” IEEE Trans. Neural Netw. Learn. Syst., vol. 36, no. 8, pp. 14937−14948, Aug. 2025. doi: 10.1109/TNNLS.2025.3527581
    [43]
    X. Zhou, J. Wang, E. Sui, W. Wang, and J. Li, “Multidimensional attention-based GAIN for missing data imputation in multimode batch processes,” IEEE Trans. Instrum. Meas., vol. 74, Art. no. 2544812, Aug. 2025. doi: 10.1109/tim.2025.3602575
    [44]
    Z. Yao and C. Zhao, “FIGAN: A missing industrial data imputation method customized for soft sensor application,” IEEE Trans. Autom. Sci. Eng., vol. 19, no. 4, pp. 3712−3722, Oct. 2022. doi: 10.1109/TASE.2021.3132037
    [45]
    Z. Zhan, D. Ma, X. Hu, and S. Zhang, “An online collaborative imputation method for industrial missing data based on multiscale MATGAN in edge computing,” IEEE Internet Things J., vol. 12, no. 10, pp. 14244−14253, May 2025. doi: 10.1109/JIOT.2025.3525815
    [46]
    B. Lu, Q. Miao, Y. Liu, T. S. Tamir, H. Zhao, X. Zhang, Y. Lv, and F.-Y. Wang, “A diffusion model for traffic data imputation,” IEEE/CAA J. Autom. Sinica, vol. 12, no. 3, pp. 606−617, Mar. 2025. doi: 10.1109/JAS.2024.124611
    [47]
    H. Xu, Z. Liu, H. Wang, C. Li, Y. Niu, W. Wang, and X. Liu, “Denoising diffusion straightforward models for energy conversion monitoring data imputation,” IEEE Trans. Ind. Informat., vol. 20, no. 10, pp. 11987−11997, Oct. 2024. doi: 10.1109/TII.2024.3413349
    [48]
    D. Liu, Y. Wang, C. Liu, X. Yuan, K. Wang, and C. Yang, “Scope-free global multi-condition-aware industrial missing data imputation framework via diffusion transformer,” IEEE Trans. Knowl. Data Eng., vol. 36, no. 11, pp. 6977−6988, Nov. 2024. doi: 10.1109/TKDE.2024.3392897
    [49]
    H. Zhao, K. Sun, A. Dezfouli, and E. V. Bonilla, “Transformed distribution matching for missing value imputation,” in Proc. 40th Int. Conf. Machine Learning, Honolulu, USA, 2023, pp. 42159−42186.
    [50]
    B. Muzellec, J. Josse, C. Boyer, and M. Cuturi, “Missing data imputation using optimal transport,” in Proc. 37th Int. Conf. Machine Learning, 2020, pp. 7130−7140.
    [51]
    G. Monge, “Memoire sur la theorie des deblais et des remblais,” Mem. Math. Phys. Acad. Royale Sci., pp. 666−704, 1781.
    [52]
    L. V. Kantorovich, “On the translocation of masses,” J. Math. Sci., vol. 133, no. 4, pp. 1381−1382, Mar. 2006. doi: 10.1287/mnsc.5.1.1
    [53]
    Y. Disser and M. Skutella, “The simplex algorithm is NP-mighty,” ACM Trans. Algorithms, vol. 15, no. 1, Art. no. 5, Jan. 2019. doi: 10.1137/1.9781611973730.59
    [54]
    K. Fatras, T. Séjourné, R. Flamary, and N. Courty, “Unbalanced minibatch optimal transport; applications to domain adaptation,” in Proc. 38th Int. Conf. Machine Learning, 2021, pp. 3186−3197.
    [55]
    L. Chizat, G. Peyré, B. Schmitzer, and F.-X. Vialard, “Scaling algorithms for unbalanced optimal transport problems,” Math. Comput., vol. 87, no. 314, pp. 2563−2609, Feb. 2018. doi: 10.1090/mcom/3303
    [56]
    T. Séjourné, J. Feydy, F.-X. Vialard, A. Trouvé, and G. Peyré, “Sinkhorn divergences for unbalanced optimal transport,” arXiv preprint arXiv: 1910.12958, 2019.
    [57]
    J. Altschuler, J. Weed, and P. Rigollet, “Near-linear time approximation algorithms for optimal transport via Sinkhorn iteration,” in Proc. 31st Int. Conf. Neural Information Processing Systems, Long Beach, USA, 2017, pp. 1961−1971.
    [58]
    P. E. Dvurechensky, A. V. Gasnikov, and A. Kroshnin, “Computational optimal transport: Complexity by accelerated gradient descent is better than by Sinkhorn’s algorithm,” in Proc. 35th Int. Conf. Machine Learning, Stockholm, Sweden, 2018, pp. 1366−1375.
    [59]
    X. Yuan, B. Huang, Y. Wang, C. Yang, and W. Gui, “Deep learning-based feature representation and its application for soft sensor modeling with variable-wise weighted SAE,” IEEE Trans. Ind. Informat., vol. 14, no. 7, pp. 3235−3243, Jul. 2018. doi: 10.1109/TII.2018.2809730
    [60]
    D. P. Kingma and J. Ba, “Adam: A method for stochastic optimization,” in Proc. 3rd Int. Conf. Learning Representations, San Diego, USA, 2015, pp. 1−9.
    [61]
    L. Yao and Z. Ge, “Industrial big data modeling and monitoring framework for plant-wide processes,” IEEE Trans. Ind. Informat., vol. 17, no. 9, pp. 6399−6408, Sep. 2021. doi: 10.1109/TII.2020.3010562
    [62]
    G. Peyré and M. Cuturi, Computational Optimal Transport, Now Publishers Inc., 2019.
    [63]
    N. Courty, R. Flamary, D. Tuia, and A. Rakotomamonjy, “Optimal transport for domain adaptation,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 39, no. 9, pp. 1853−1865, Sep. 2017. doi: 10.1109/TPAMI.2016.2615921
    [64]
    N. Bonneel, M. van de Panne, S. Paris, and W. Heidrich, “Displacement interpolation using Lagrangian mass transport,” in Proc. 2011 SIGGRAPH Asia Conf., Hong Kong, China, 2011, Art. no. 158.
    [65]
    H. Wang, Z. Chen, Z. Liu, X. Chen, H. Li, and Z. Lin, “Proximity matters: Local proximity enhanced balancing for treatment effect estimation,” Proc. 31st ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining V.2, Toronto, Canada, 2025, pp. 2927−2937.
    [66]
    H. Wang, Z. Chen, H. Zhang, Z. Li, L. Pan, H. Li, and M. Gong, “Debiased recommendation via Wasserstein causal balancing,” ACM Trans. Inf. Syst., vol. 43, no. 6, Art. no. 145, Nov. 2025. doi: 10.1145/3725731
    [67]
    S. D. Marino and A. Gerolin, “An optimal transport approach for the schrödinger bridge problem and convergence of Sinkhorn algorithm,” J. Sci. Comput., vol. 85, no. 2, Art. no. 27, Oct. 2020. doi: 10.1007/s10915-020-01325-7
    [68]
    H. Xu, D. Luo, H. Zha, and L. Carin, “Gromov-Wasserstein learning for graph matching and node embedding,” in Proc. 36th Int. Conf. Machine Learning, Long Beach, USA, 2019, pp. 6932−6941.
    [69]
    H. Wang, Z. Chen, J. Fan, H. Li, T. Liu, W. Liu, Q. Dai, Y. Wang, Z. Dong, and R. Tang, “Optimal transport for treatment effect estimation,” in Proc. 37th Int. Conf. Neural Information Processing Systems, New Orleans, USA, 2023, Art. no. 237.
    [70]
    H. Wang, Z. Li, H. Li, X. Chen, M. Gong, B. Chen, and Z. Chen, “Optimal transport for time series imputation,” in Proc. 13th Int. Conf. Learning Representations, Singapore, Singapore, 2025, pp. 1−9.
    • Related Articles
    • Supplements(0)

Catalog

    通讯作者: 陈斌, bchen63@163.com
    • 1. 

      沈阳化工大学材料科学与工程学院 沈阳 110142

    1. 本站搜索
    2. 百度学术搜索
    3. 万方数据库搜索
    4. CNKI搜索

    Figures(7)  / Tables(8)

    Get Citation
    PDF
    XML

    Article Metrics

    Article views (13) PDF downloads(0) Cited by()

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return