Real-Time Decision Intelligence Using Adaptive Deep Learning
DOI:
https://doi.org/10.5281/zenodo.20591830Keywords:
Adaptive deep learning,Real-time decision support,Dynamic business environments,Online learning,Continual learning,Concept drift detection,Streaming data analytics,Edge AI inference,Low-latency prediction,Reinforcement learning for business,Time-series forecasting,Anomaly detection,Automated model retraining,MLOps for real-time systems,Explainable AI (XAI) in decision support.Abstract
Adaptive deep learning models are needed to support real-time operations in dynamic business environments characterized by fast state changes. Streaming data processing architectures, model adaptation mechanisms, and decision-trust capabilities form the building blocks for adaptive architectures. Adaptation relies on a spectrum of techniques, from online learning and continual learning through meta-learning to lightweight and edge-enabled neural networks. Evaluation is also challenging: approaches must be benchmarked in dynamic non-stationary settings using appropriate real-time metrics. In addition, vigilance is needed to ensure robustness, fairness, and compliance in controlled areas, such as financial services and healthcare. Real-time adaptive support has been achieved in practical scenarios involving supply chains, healthcare resource allocation, and financial-services operations.
References
[1] Hinder, F., Artelt, A., Hammer, B., & Biehl, M. (2024). Concept drift: A survey on detecting drifts in evolving environments. Frontiers in Artificial Intelligence, 7, Article 1378445.
[2] Wang, L., Zhang, Y., Li, H., & Yao, Y. (2023). A comprehensive survey of continual learning: Theory, method and application. arXiv.
[3] Leo, J., & Madhu, B. (2024). Survey of continuous deep learning methods and incremental learning techniques. Neurocomputing, 575, 127105.
[4] Shyaa, M. A., Al-Jarrah, O. Y., & Yoo, P. D. (2024). Evolving cybersecurity frontiers: A comprehensive survey on concept drift and feature dynamics aware machine and deep learning in intrusion detection systems. Engineering Applications of Artificial Intelligence, 135, 109143.
[5] Gunasekara, N., & Krempl, G. (2024). Recurrent concept drifts on data streams. In Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence (IJCAI 2024) (pp. 1–9).
[6] Rodrigues, M. G., et al. (2025). A MLOps architecture for near real-time distributed stream learning. Journal of Network and Computer Applications, 229, 103873.
[7] Deng, Z., et al. (2025). Achieving trustworthy real-time decision support systems with low-latency AI models. arXiv.
[8] Gepperth, A. (2024). Continual improvement of deep neural networks in the real world. In Proceedings of ESANN 2024 (pp. 1–8).
[9] Li, A., Li, H., & Yuan, G. (2024). Continual learning with deep neural networks in physiological signal data: A survey. Healthcare, 12(2), 155.
[10] Yan, Y., Wang, Z., & Liu, Q. (2024). Big data analysis and decision support system based on deep learning: A marketing decision perspective. Computer-Aided Design, 21(S13), 62–74.
[11] Sobrie, L., et al. (2024). Real-time decision support for human–machine interaction in safety-critical environments using machine learning. Decision Support Systems, 179, 114198.
[12] Yang, S., et al. (2025). A survey of processing methods for different types of concept drift in streaming data. Decision Support Systems, 184, 114356.
[13] Chitraju, S. (2024). End-to-end ML operations (MLOps): Enhancing model reliability and performance at scale. SSRN Electronic Journal.
[14] Gama, J., Žliobaitė, I., Bifet, A., Pechenizkiy, M., & Bouchachia, A. (2014). A survey on concept drift adaptation. ACM Computing Surveys, 46(4), 44.
[15] Webb, G. I., Hyde, R., Cao, H., Nguyen, H. L., & Petitjean, F. (2016). Characterizing concept drift. Data Mining and Knowledge Discovery, 30(4), 964–994.
[16] Cerqueira, V., Torgo, L., Soares, C., & Harrison, P. (2020). Machine learning for time series forecasting: A survey. ACM Computing Surveys, 54(6), 1–36.
[17] Lim, B., Arık, S. Ö., Loeff, N., & Pfister, T. (2021). Temporal fusion transformers for interpretable multi-horizon time series forecasting. International Journal of Forecasting, 37(4), 1748–1764.
[18] Oreshkin, B. N., Carpov, D., Chapados, N., & Bengio, Y. (2020). N-BEATS: Neural basis expansion analysis for interpretable time series forecasting. International Conference on Learning Representations.
[19] Zhou, D., et al. (2021). Informer: Beyond efficient transformer for long sequence time-series forecasting. In Proceedings of AAAI (pp. 11106–11115).
[20] Zerveas, G., et al. (2021). A transformer-based framework for multivariate time series representation learning. In Proceedings of KDD (pp. 2114–2124).
[21] Ruff, L., et al. (2021). Unifying review of deep and shallow anomaly detection. Proceedings of the IEEE, 109(5), 756–795.
[22] Pang, G., Shen, C., Cao, L., & Hengel, A. V. D. (2021). Deep learning for anomaly detection: A review. ACM Computing Surveys, 54(2), 1–38.
[23] Sculley, D., Holt, G., Golovin, D., Davydov, E., Phillips, T., Ebner, D., Chaudhary, V., Young, M., Crespo, J.-F., & Dennison, D. (2015). Hidden technical debt in machine learning systems. In Advances in Neural Information Processing Systems (pp. 2503–2511).
[24] Amershi, S., Begel, A., Bird, C., DeLine, R., Gall, H., Kamar, E., Nagappan, N., Nushi, B., Zimmermann, T., & others. (2019). Software engineering for machine learning: A case study. In Proceedings of ICSE (pp. 291–300).
[25] Huyen, C. (2022). Designing machine learning systems: An iterative process for production-ready applications. O’Reilly Media.
[26] Breck, E., Cai, S., Nielsen, E., Salib, M., & Sculley, D. (2017). The ML test score: A rubric for ML production readiness and technical debt reduction. In Proceedings of IEEE Big Data (pp. 1123–1132).
[27] Ribeiro, M. T., Singh, S., & Guestrin, C. (2016). “Why should I trust you?” Explaining the predictions of any classifier. In Proceedings of KDD (pp. 1135–1144).
[28] Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. In Advances in Neural Information Processing Systems (pp. 4765–4774).
[29] Molnar, C. (2022). Interpretable machine learning: A guide for making black box models explainable (2nd ed.). Leanpub.
[30] Doshi-Velez, F., & Kim, B. (2017). Towards a rigorous science of interpretable machine learning. arXiv.
[31] Sutton, R. S., & Barto, A. G. (2018). Reinforcement learning: An introduction (2nd ed.). MIT Press.
[32] Mnih, V., et al. (2015). Human-level control through deep reinforcement learning. Nature, 518(7540), 529–533.
[33] Haarnoja, T., Zhou, A., Abbeel, P., & Levine, S. (2018). Soft actor-critic: Off-policy maximum entropy deep reinforcement learning with a stochastic actor. In Proceedings of ICML (pp. 1861–1870).
[34] Silver, D., et al. (2018). A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play. Science, 362(6419), 1140–1144.
[35] Gomes, H. M., Read, J., Bifet, A., Barddal, J. P., & Gama, J. (2019). Machine learning for streaming data: State of the art, challenges, and opportunities. SIGKDD Explorations, 21(2), 6–22.
[36] Bifet, A., & Gavalda, R. (2007). Learning from time-changing data with adaptive windowing. In Proceedings of SIAM ICDM (pp. 443–448).
[37] Baena-García, M., del Campo-Ávila, J., Fidalgo, R., Bifet, A., Gavaldà, R., & Morales-Bueno, R. (2006). Early drift detection method. In Proceedings of the 4th International Workshop on Knowledge Discovery from Data Streams (pp. 77–86).
[38] Montiel, J., Read, J., Bifet, A., & Abdessalem, T. (2018). Scikit-multiflow: A multi-output streaming framework. Journal of Machine Learning Research, 19(72), 1–5.
[39] Polyzotis, N., Roy, S., Whang, S. E., & Zinkevich, M. (2018). Data management challenges in production machine learning. In Proceedings of SIGMOD (pp. 1723–1726).
[40] Kumar, A., et al. (2020). Model selection and hyperparameter tuning for real-time machine learning systems: Challenges and best practices. Proceedings of the VLDB Endowment, 13(12), 3163–3176.
Additional Files
Published
Data Availability Statement
NONE
