Hybrid Deep-Ensemble Network with VAE-Based Augmentation for Imbalanced Tabular Data Classification
Source abstract
Abstract
Background: Severe class imbalance limits reliable tabular AI in manufacturing, finance, and healthcare. Methods: We built a modular pipeline comprising correlation-aware seriation; a hybrid convolutional neural network (CNN)–transformer–Bidirectional Long Short-Term Memory (BiLSTM) encoder; variational autoencoder (VAE)-based minority augmentation; and deep/tree ensemble heads (XGBoost and Support Vector Machine, SVM). We benchmarked the Synthetic Minority Oversampling Technique (SMOTE) and ADASYN under identical protocols. Focal loss and ensemble weights were tuned per dataset. The primary metric was the Area Under the Precision–Recall Curve (AUPRC), with receiver operating characteristic area under the curve (ROC AUC) as complementary. Synthetic-data fidelity was quantified by train-on-synthetic/test-on-real (TSTR) utility, two-sample discriminability (ROC AUC of a real-vs-synthetic classifier), and Maximum Mean Discrepancy (MMD2). Results: Across five datasets (SECOM, CREDIT, THYROID, APS, and UCI), augmentation was data-dependent: VAE led on APS (+3.66 pp AUPRC vs. SMOTE) and was competitive on CREDIT (+0.10 pp vs. None); the SMOTE dominated SECOM; no augmentation performed best for THYROID and UCI. Positional embedding (PE) with seriation helped when strong local correlations were present. Ensembles typically favored XGBoost while benefiting from the hybrid encoder. Efficiency profiling and a slim variant supported latency-sensitive use. Conclusions: A data-aware recipe emerged: prefer VAE when fidelity is high, the SMOTE on smoother minority manifolds, and no augmentation when baselines suffice; apply PE/seriation selectively and tune per dataset for robust, reproducible deployment.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Imbalanced Data Classification Techniques · Vehicle License Plate Recognition · Electricity Theft Detection Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Multimodal AX Business Team, LG CNS Co., Ltd., Seoul 07795, Republic of Korea
Department of Computer Engineering, Tech University of Korea, Siheung 15073, Republic of Korea
Funding
Korean government (MSIT)
References (35)
Show source-provided references
1. Learning from Imbalanced Data10.1109/TKDE.2008.239 ↗
2. Survey on Deep Learning with Class Imbalance10.1186/s40537-019-0192-5 ↗
3. SMOTE: Synthetic Minority Over-Sampling Technique10.1613/jair.953 ↗
4. He, H., Bai, Y., Garcia, E.A., and Li, S. (2008, January 1–8). ADASYN: Adaptive Synthetic Sampling Approach for Imbalanced Learning. Proceedings of the 2008 IEEE International Joint Conference on Neural Networks (IJCNN), Hong Kong, China.
5. Kingma, D.P., and Welling, M. (2014). Auto-encoding Variational Bayes. arXiv.
6. Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., and Bengio, Y. (2014, January 8–13). Generative Adversarial Nets. Proceedings of the Advances in Neural Information Processing Systems 27 (NeurIPS 2014), Montreal, QC, Canada.
7. Esteban, C., Hyland, S.L., and Rätsch, G. (2017). Real-Valued (Medical) Time Series Generation with Recurrent Conditional GANs. arXiv.
8. Xu, L., Skoularidou, M., Cuesta-Infante, A., and Veeramachaneni, K. (2019). Modeling Tabular Data Using Conditional GAN. arXiv.
9. Support-Vector Networks10.1023/A:1022627411411 ↗
10. Random Forests10.1023/A:1010933404324 ↗
11. Kotelnikov, A., Baranchuk, D., Rubachev, I., and Babenko, A. (2023, January 10–16). TabDDPM: Modeling Tabular Data with Diffusion Models. Proceedings of the Advances in Neural Information Processing Systems (NeurIPS 2023), New Orleans, LA, USA.
12. Gradient-Based Learning Applied to Document Recognition10.1109/5.726791 ↗
13. Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, Ł., and Polosukhin, I. (2017, January 4–9). Attention Is All You Need. Proceedings of the Advances in Neural Information Processing Systems 30 (NeurIPS 2017), Long Beach, CA, USA.
14. Chen, T., and Guestrin, C. (2016, January 13–17). XGBoost: A Scalable Tree Boosting System. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD ’16), San Francisco, CA, USA.10.1145/2939672.2939785 ↗
15. Long Short-Term Memory10.1162/neco.1997.9.8.1735 ↗
16. A Kernel Two-Sample Test
17. UCI—SECOM Dataset (2025, September 18). UCI Machine Learning Repository. Available online: https://archive.ics.uci.edu/dataset/179/secom.
18. Kaggle (2025, September 18). Credit Card Fraud Detection Dataset. Available online: https://www.kaggle.com/mlg-ulb/creditcardfraud.
19. Rayana, S. (2016). Thyroid Disease (ODDS). Outlier Detection DataSets (ODDS), Stony Brook University. Available online: https://shebuti.com/thyroid-disease-dataset/.
20. UCI—APS Failure at Scania Trucks (2025, September 18). UCI Machine Learning Repository. Available online: https://www.kaggle.com/datasets/uciml/aps-failure-at-scania-trucks-data-set.
21. UCI—Default of Credit Card Clients (2025, September 18). UCI Machine Learning Repository. Available online: https://archive.ics.uci.edu/dataset/350/default+of+credit+card+clients.
22. Huang, X., Khetan, A., Cvitkovic, M., and Karnin, Z. (2020). TabTransformer: Tabular Data Modeling Using Contextual Embeddings. arXiv.
23. Bidirectional Recurrent Neural Networks10.1109/78.650093 ↗
24. Gorishniy, Y., Rubachev, I., Khrulkov, V., and Babenko, A. (2021, January 6–14). Revisiting Deep Learning Models for Tabular Data. Proceedings of the Advances in Neural Information Processing Systems (NeurIPS 2021), Virtual.
25. Lin, T.-Y., Goyal, P., Girshick, R., He, K., and Dollár, P. (2017, January 22–29). Focal Loss for Dense Object Detection. Proceedings of the IEEE International Conference on Computer Vision (ICCV 2017), Venice, Italy.10.1109/ICCV.2017.324 ↗
26. The Precision–Recall Plot Is More Informative than the ROC Plot When Evaluating Binary Classifiers on Imbalanced Datasets10.1371/journal.pone.0118432 ↗
27. Zhang, H., Cisse, M., Dauphin, Y.N., and Lopez-Paz, D. (May, January 30). mixup: Beyond Empirical Risk Minimization. Proceedings of the International Conference on Learning Representations (ICLR 2018), Vancouver, BC, Canada.
28. Yun, S., Han, D., Oh, S.J., Chun, S., Choe, J., and Yoo, Y. (November, January 27). CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features. Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV 2019), Seoul, Republic of Korea.10.1109/ICCV.2019.00612 ↗
29. PointCutMix: Regularization Strategy for Point Clouds
30. Prokhorenkova, L., Gusev, G., Vorobev, A., Dorogush, A.V., and Gulin, A. (2018, January 3–8). CatBoost: Unbiased Boosting with Categorical Features. Proceedings of the 2018 Conference on Neural Information Processing Systems (NeurIPS 2018), Montreal, QC, Canada.
31. TabNet: Attentive Interpretable Tabular Learning
32. Somepalli, G., Goldblum, M., Schwarzschild, A., Bruss, C.B., and Goldstein, T. (2021). SAINT: Improved Neural Networks for Tabular Data via Row Attention and Contrastive Pre-Training. arXiv.
33. Hollmann, N., Müller, S., Eggensperger, K., and Hutter, F. (December, January 28). TabPFN: A Transformer That Solves Small Tabular Classification Problems in a Second. Proceedings of the 2022 Conference on Neural Information Processing Systems (NeurIPS 2022), New Orleans, LA, USA.
34. Radicioni, L., Bono, F.M., and Cinquemani, S. (2025). Vibration-Based Anomaly Detection in Industrial Machines: A Comparison of Autoencoders and Latent Spaces. Machines, 13.10.3390/machines13020139 ↗
35. Alsaif, M., Alaqel, H., Almalaq, A., Alharbi, R., Alshahrani, M., and Pasha, M. (2022). A Novel Data Augmentation-Based Brain Tumor Detection Using Convolutional Neural Network. Appl. Sci., 12.10.3390/app12083773 ↗