Open Access
Research Article
Approach of VAEs in handling incomplete heterogeneous data
Ari Ernesto Ortiz Castellanost107999401@ntut.org.tw; aryernesto@hotmail.comCollege of Electrical Engineering and Computer Science National Taipei University of Technology 1, Sec. 3, Zhongxiao E. Rd.Taipei, 10608, Taiwan (R.O.C.)View full profile → , *Chuan-Ming LiuCorresponding authorcmliu@ntut.edu.twDepartment of Computer Science and Information Engineering National Taipei University of Technology 1, Sec. 3, Zhongxiao E. Rd.Taipei, 10608, Taiwan (R.O.C.)View full profile →
* Corresponding author · click or hover a name for details
- Received:
- 01 Nov 2024
- Published Online:
- 01 Jun 2026
- Article type:
- Research Article
- Language:
- EN
- Article no.:
- JIOS-1963
- Pages:
- 2139–2153
Abstract
The latent structure found in vast, intricate, high-dimensional datasets is effectively and accurately captured by Variational autoencoders (VAEs). Existing VAE architectures struggle with issues commonly found in real-world datasets: heterogeneous data (which combines continuous and discrete variables) and incomplete data (where missing values appear randomly). This paper introduces a comprehensive framework for designing VAEs capable of accommodating incomplete heterogeneous data. Termed MI-VAE, this novel approach utilizes a key feature: likelihood models tailored for diverse data types, including real-valued, positive real-valued, interval, categorical, ordinal, and count data. Furthermore, the model is proficient at managing missing values, accurately performing both estimation and imputation functions. Moreover, when applied to supervised learning problems, MI-VAE demonstrates a competitive predictive capacity, often exceeding that of supervised models built from the same incomplete data. Furthermore, MI-VAE not only addresses missing values but also achieves competitive performance in supervised learning tasks, demonstrating its effectiveness in real-world applications.
Keywords
Subject Classifications
Incomplete dataVariational encodersVAEsFrameworks
References
[1] T. Sudak and S. Tschiatschek, “Posterior consistency for missing data in variational autoencoders,” in Joint European Conference on Machine Learning and Knowledge Discovery in Databases, pp. 508-524 (2023).[2] B. Smith, S. V. Van Steelandt, and A. Khojandi, “Evaluating the impact of healthcare data completeness for deep generative models,” Methods of Information in Medicine, vol. 62, no. 1-2, pp. 31-39 (2023), doi: 10.1055/a-2023-9181.[3] B. Roskams-Hieter, J. Wells, and S. Wade, “Leveraging variational autoencoders for multiple data imputation,” in Machine Learning and Knowledge Discovery in Databases: Research Track, pp. 491-506 (2023).[4] V. Simkus, B. Rhodes, and M. U. Gutmann, “Variational Gibbs inference for statistical model estimation from incomplete data,” Journal of Machine Learning Research, vol. 24, pp. 1-72 (2023).[5] C. Zhao, K.-J. Su, C. Wu, X. Cao, Q. Sha, W. Li, Z. Luo, T. Qin, C. Qiu, L. J. Zhao, A. Liu, L. Jiang, X. Zhang, H. Shen, W. Zhou, and H.-W. Deng, “Multi-view variational autoencoder for missing value imputation in untargeted metabolomics,” PubMed Central (2023).[6] Q. Ma, X. Li, B. Mei, X. Wang, B. Ning, and G. Li, “MI-VAE: Multiple imputation based on variational auto-encoder,” Engineering Applications of Artificial Intelligence, vol. 123 (2023).[7] D. Papadopoulos and V. D. Karalis, “Variational autoencoders for data augmentation in clinical studies,” Applied Sciences, vol. 13, no. 15, p. 8793 (2023), doi: 10.3390/app13158793.[8] J. T. McCoy, S. Kroon, and L. Auret, “Variational autoencoders for missing data imputation with application to a simulated milling circuit,” IFAC-PapersOnLine, pp. 141-146 (2018), doi: 10.1016/j.ifacol.2018.09.406.[9] B. Jafrasteh, D. Hernandez-Lobado, S. P. Lucian-Lopez, and I. Benavente-Fernandez, “Gaussian processes for missing value imputation,” Knowledge-Based Systems, vol. 273 (2023), doi: 10.1016/j.knosys.2023.110603.[10] M. Lichman, “UCI machine learning repository,” (2013). [Online]. Available: http://archive.ics.uci.edu/ml
Views: 104Downloads: 27Citations: 0
