Abstract:
Missing-data imputation has become a necessary task to address as missing data problems have become prevalent in modern datasets, and their presence can be detrimental to statistical inference, machine-learning algorithms, and predictive performance. State-of-the-art imputation approaches often lack flexibility in the modeling of nonlinear relationships and dependencies. This provides motivation for deep generative approaches that can first learn the underlying data distribution to generate candidate imputations for missing observations. This survey covers recent advances in deep generative missing-data imputation approaches, including autoencoders, GANs, diffusion approaches, and related attention, graph, and hybrid models. We provide an overview of studies on multiple datasets, including healthcare, time-series, spatiotemporal, educational, and genomic datasets, and compare methodologies based on modeling rationale, merits and setbacks, and downstream prediction implications. The results demonstrate that diffusion based approaches are promising for temporal and spatiotemporal contexts, VAEs offer flexible probabilistic representations, and GANs excel at distribution matching, but computational efficiency, accounting for uncertainty, MNAR mechanisms, scalability, privacy preservation, and predictive performance limitations persist. The study highlights directions for future methodology, including prediction-centric imputation, scalable models, and privacy protection.
Keywords: Missing Data, Data Imputation, Deep Learning, Generative Models, Variational Autoencoder, Generative Adversarial Network, Diffusion Models, Time Series, Prediction.
REFERENCES
Abdullah, A.A. and Abdullah, D.B. (2025) “A Unified Framework for Real-Time Intrusion Detection and Mitigation using Apache Spark,” Journal of Wireless Mobile Networks, Ubiquitous Computing, and Dependable Applications, 16(4), pp. 642–662. Available at: https://doi.org/10.58346/JOWUA.2025.I4.037.
Afkanpour, M., Hosseinzadeh, E. and Tabesh, H. (2024) “Identify the most appropriate imputation method for handling missing values in clinical structured datasets: a systematic review,” BMC Medical Research Methodology, 24(1). Available at: https://doi.org/10.1186/s12874-024-02310-6.
Albasheer, K.D. and Abdullah, D.B. (2025) “Fog and edge computing and its role in distributed real- time containers: A survey,” AIP Conference Proceedings, p. 030009. Available at: https://doi.org/10.1063/5.0286172.
Alcaraz, J.M.L. and Strodthoff, N. (2023) “Diffusion-based Time Series Imputation and Forecasting with Structured State Space Models,” Transactions on Machine Learning Research, 2023- Febru, pp. 1–36.
Buwaneswaran, M. (2024) “Scholarship @ Western Missing Data Imputation for Smart Meters : Conditional Denoising Diffusion Model & Temporally Chained Equations Keywords.”
Chen, J., Xu, Y., Wang, P.and Yang Y. (2023) “Deep Generative Imputation Model for Missing Not At Random Data,” International Conference on Information and Knowledge Management, Proceedings, pp. 316–325. Available at: https://doi.org/10.1145/3583780.3614835.
Chen, Z., Li, H., Wang, F., Zhang, O., Xu, H., Jiang, X., Song, Z. and Wang, E. (2024) “Rethinking the Diffusion Models for Numerical Tabular Data Imputation from the Perspective of Wasserstein Gradient Flow,” (NeurIPS), pp. 1–54. Available at: http://arxiv.org/abs/2406.15762.
Ghanavi, R., Liang, B. and Tizghadam, A. (2021) “Generative Adversarial Classification Network with Application to Network Traffic Classification,” Proceedings – IEEE Global Communications Conference, GLOBECOM [Preprint]. Available at: https://doi.org/10.1109/GLOBECOM46510.2021.96 85899.
Ghosh, S., Nolte, D., Tsurukawa, F. and Pal R. (2026) “Federated generative imputation with variational and denoising autoencoders for pan- cancer genomics,” Heliyon, 12(4), p. e44690. Available at: https://doi.org/10.1016/j.heliyon.2026.e44690.
Hallaji, E., Razavi-Far, R., Palade, V.and Saif , M. (2021) “Adversarial Learning on Incomplete and Imbalanced Medical Data for Robust Survival Prediction of Liver Transplant Patients,” IEEE Access, 9, pp. 73641–73650. Available at: https://doi.org/10.1109/ACCESS.2021.3081040.
Kazijevs, M. and Samad, M.D. (2023) “Deep imputation of missing values in time series health data: A review with benchmarking,” Journal of Biomedical Informatics, 144(Mvi). Available at: https://doi.org/10.1016/j.jbi.2023.104440.
Kowsar, I. (2025) “Digital Scholarship @ Tennessee State University DEEP IMPUTATION OF MISSING VALUES USING FEATURE AND.”
Kuang, S., Huang, Y. and Song, J. (2025) “Unsupervised data imputation with multiple importance sampling variational autoencoders,” Scientific Reports, 15(1), pp. 1–16. Available at: https://doi.org/10.1038/s41598-025-87641-0.
Li, X. (2024) “Time Series Forecasting with Missing Data Using Generative Adversarial Networks and Bayesian Inference,” Information (Switzerland), 15(4). Available at: https://doi.org/10.3390/info15040222.
Liu, M., Huang, H., Feng, H., Sun, L., Du, B.and Fu Y (2023) “PriSTI: A Conditional Diffusion Framework for Spatiotemporal Imputation,” Proceedings – International Conference on Data Engineering, 2023-April, pp. 1927–1939. Available at: https://doi.org/10.1109/ICDE55515.2023.00150.
Matrood, H.H. (2026) “Journal of Economics, Business, and Commerce (JEBC),” 3(1), pp. 226– 235. Available at: https://doi.org/10.69739/jebc.v3i1.1980
Mattei, P.A. and Freiisen, J. (2019) “Miwae: Deep generative modelling and imputation of incomplete data sets,” 36th International Conference on Machine Learning, ICML 2019, 2019-June, pp. 7762–7772.
Najeeb, R.S.M. and Dahl, I.O.A.M. (2022) “Brain Tumor Segmentation Utilizing Generative Adversarial, Resnet And Unet Deep Learning,” 2022 8th International Conference on Contemporary Information Technology and Mathematics, ICCITM 2022, pp. 85–89. Available at: https://doi.org/10.1109/ICCITM56309.2022.100317 60.
Pereira, R.C., Abreu, P., Rodrigues, P. and Figueiredo M. (2024) “Imputation of data Missing Not at Random: Artificial generation and benchmark analysis,” Expert Systems with Applications, 249(PB), p. 123654. Available at: https://doi.org/10.1016/j.eswa.2024.123654.
Qasha, R. and Sulyaman, H. (2025) “Microservices caching for container-based IoT system in the edge and cloud,” Indonesian Journal of Electrical Engineering and Computer Science, 39(3), p. 1652. Available at: https://doi.org/10.11591/ijeecs.v39.i3.pp1652-1660.
Sultan, N.A. and Abdullah, D.B. (2023)“Investigating scientific collaboration networks in Iraq using cloud computing and data mining,” International Journal of Applied Science and Engineering, 20(4). Available at: https://doi.org/10.6703/IJASE.202312_20(4).003.
Tashiro, Y. ,Song, J., Song, Y. and Ermon S. (2021) “CSDI: Conditional Score-based Diffusion Models for Probabilistic Time Series Imputation,” Advances in Neural Information Processing Systems, 30(NeurIPS), pp. 24804–24816. Available at: https://doi.org/10.5555/3540261.3542161.
Veldkamp, K., Grasman, R. and Molenaar, D. (2025) “Handling missing data in variational autoencoder based item response theory,” British Journal of Mathematical and Statistical Psychology, 78(1), pp. 378–397. Available at: https://doi.org/10.1111/bmsp.12363.
Wang, D., Yan, Y., Qiu, R., Zhu, Y., Guan, K., Margenot, A. and Tong, H. (2023) “Networked Time Series Imputation via Position-aware Graph Enhanced Variational Autoencoders,” Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 2256– 2268. Available at: https://doi.org/10.1145/3580305.3599444.
Wang, X. , Zhang, H., Wang, P., Zhang, Y., Wang, B., Zhou, Z. and Wang Y. (2023) “An Observed Value Consistent Diffusion Model for Imputing Missing Values in Multivariate Time Series,” Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 2409–2418. Available at: https://doi.org/10.1145/3580305.3599257.
Xiao, C. ,Jiang, X., Du, X., Yang, W., Lu, W., Wang, X. and Chetty, K. (2025) “Boundary-enhanced time series data imputation with long-term dependency diffusion models,” Knowledge-Based Systems, 310, pp. 1–17. Available at: https://doi.org/10.1016/j.knosys.2024.112917.
Xu, X. ,Hu, L., Liu, J., Miao, X. and Yan X. (2026) “T-LSTM-VAE: A random missing data imputation model for multivariate time series,” Journal of the Franklin Institute, 363(1), p. 108231. Available at: https://doi.org/10.1016/j.jfranklin.2025.108231.
Yoon, J., Jordon, J. and Van Der Schaar, M. (2018) “Supplementary materials GAIN: Missing data imputation using generative adversarial nets,”35th International Conference on Machine Learning, ICML 2018, 13(1), pp. 9052–9059.
Zhang, L., Lin, J., Sabatini, J., Zapata-Rivera, D., Forsyth, C.,Jiang, Y., Hollander, J., Hu, X. and Graesser A. (2025) “Generative Data Imputation for Sparse Learner Performance Data Using Generative Adversarial Imputation Networks,” 18(9), pp. 1–17. Available at: https://arxiv.org/pdf/2503.18982.
Zhang, Y. (2024) “Missing Data Imputation for Health Care Big Data using Denoising Autoencoder with Generative Adversarial Network,” Scalable Computing: Practice and Experience, 25(5), pp. 3850–3857. Available at: https://doi.org/10.12694/scpe.v25i5.3023.