Model Monitoring Frameworks in Data Science and AI: Ensuring Robustness, Fairness, and Continuous Improvement

Praneeth Reddy Amudala Puchakayala

Authors

Praneeth Reddy Amudala Puchakayala Data scientist, Regions Bank. Author

Keywords:

Artificial Intelligence, Machine Learning, Data drift, Model Drift, Banking, Accuracy

Abstract

Our study's findings highlight the importance of AI/ML systems in banks and point to a growing trend of banks integrating these technologies more thoroughly. On the other hand, Vol. presented us with further difficulties as we explored. Based on the certain real time applications such as, Industry, Banking, etc., Various AI/ML monitoring frameworks were discussed to ensure robustness, fairness, explainability, and continuous improvement for AI models. To measure the AI/ML, Shapley values, AUROC and F1-score metrics were discussed. There the monitoring models are discussed based on the challenges such as, Model drift, Data drift, customer compliance, explainable, fariness and compliance. Solutions focused on AML regulations to monitor the customer transaction in banking sector and helps to identify and mitigate based on proactive monitoring framework. Pre-deployment, Deployment and Post-Deployment process were discussed in the model monitoring system. Finally, Key metrics such as accuracy, fairness, and explainability related to data and model drift towards detection and mitigation.

References

G. Ditzler, M. Roveri, C. Alippi, and R. Polikar, “Learning in nonstationary environments: A survey,” Comp. Intell. Mag., vol. 10, no. 4, p. 12–25, nov 2015. [Online]. Available: https: //doi.org/10.1109/MCI.2015.2471196

E. Breck, S. Cai, E. Nielsen, M. Salib, and D. Sculley, “The ml test score: A rubric for ml production readiness and technical debt reduction,” in 2017 IEEE International Conference on Big Data (Big Data), 2017, pp. 1123–1132.

C. Huyen, Designing Machine Learning Systems. O’Reilly Media, Inc., 2022. [4] S. Rabanser, S. G¨unnemann, and Z. C. Lipton, “Failing loudly: An empirical study of methods for detecting dataset shift,” in Advances in Neural Information Processing Systems 32: Annual Conference on Neural Information Processing Systems 2019, NeurIPS 2019, December 8-14, 2019,

Vancouver, BC, Canada, H. M. Wallach, H. Larochelle, A. Beygelzimer, F. d’Alch´e-Buc, E. B. Fox, and R. Garnett, Eds., 2019, pp. 1394–1406.

A. Van Looveren, J. Klaise, G. Vacanti, O. Cobb, A. Scillitoe, R. Samoilescu, and A. Athorne, “Alibi detect: Algorithms for outlier, adversarial and drift detection,” 2019. [Online]. Available: https://github.com/SeldonIO/alibi-detect

O. Cobb and A. V. Looveren, “Context-aware drift detection,” in Proceedings of the 39th International Conference on Machine Learning, 2022.

M. Abadi, A. Agarwal, P. Barham, E. Brevdo, Z. Chen, C. Citro, G. S. Corrado, A. Davis, J. Dean, M. Devin, S. Ghemawat, I. Goodfellow, A. Harp, G. Irving, M. Isard, Y. Jia, R. Jozefowicz, L. Kaiser, M. Kudlur, J. Levenberg, D. Man´e, R. Monga, S. Moore, D. Murray, C. Olah, M. Schuster, J. Shlens, B. Steiner, I. Sutskever, K. Talwar, P. Tucker, V. Vanhoucke, V. Vasudevan, F. Vi´egas, O. Vinyals, P. Warden, M. Wattenberg, M. Wicke, Y. Yu, and X. Zheng, “TensorFlow: Large-scale machine learning on heterogeneous systems,” 2015, software available from tensorflow.org. [Online]. Available: https://www.tensorflow.org/

T. Viehmann, L. Antiga, D. Cortinovis, and L. Lozza, “TorchDrift: drift detection for pytorch,” 2021. [Online]. Available: https://www. torchdrift.org/

F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V. Dubourg, J. Vanderplas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, and E. Duchesnay, “Scikit-learn: Machine learning in Python,” Journal of Machine Learning Research, vol. 12, pp. 2825–2830, 2011.

M. Zaharia, R. Xin, P. Wendell, T. Das, M. Armbrust, A. Dave, X. Meng, J. Rosen, S. Venkataraman, M. Franklin, A. Ghodsi, J. Gonzalez, S. Shenker, and I. Stoica, “Apache spark: A unified engine for big data processing,” Communications of the ACM, vol. 59, pp. 56–65, 11 2016.

J. Klaise, A. V. Looveren, C. Cox, G. Vacanti, and A. Coca, “Monitoring and explainability of models in production,” in Workshop onChallenges in Deploying and Monitoring MachineLearning Systems(ICML 2020), 2020.

G. A. Lewis, S. Echeverr´ ıa, L. Pons, and J. Chrabaszcz, “Augur: A step towards realistic drift detection in production ml systems,” in Workshop on Software Engineering for Responsible AI (SE4RAI’22 ), 2022.

M. Zaharia, A. Chen, A. Davidson, A. Ghodsi, S. A. Hong, A. Konwinski, S. Murching, T. Nykodym, P. Ogilvie, M. Parkhe, F. Xie, and C. Zumar, “Accelerating the machine learning lifecycle with mlflow,” Bulletin of the IEEE Computer Society Technical Committee on Data Engineering, 2018.

B. Chen, B. Eck, F. Fusco, R. Gormally, M. Purcell, M. Sinn, and S. Tirupathi, “Castor: Contextual IoT time series data and model management at scale,” Proc. of the 18th ICDM 2018, pp 1487-1492, 2018.

B. Eck, F. Fusco, R. Gormaly, M. Purcell, and S. Tirupathi, “Scalable deployment of AI time-series models for IoT,” in Workshop AI for Internet of Things (AI4IoT) at the 28th International Joint Conference on Artificial Intelligence (IJCAI), 2019.

W. Hummer, V. Muthusamy, T. Rausch, P. Dube, K. El Maghraoui, A. Murthi, and P. Oum, “Modelops: Cloud-based lifecycle management for reliable and trusted ai,” in 2019 IEEE International Conference on Cloud Engineering (IC2E), 2019, pp. 113–120.

“Ibm watson openscale,” accessed 1 Sept 2022. [Online]. Available: https://www.ibm.com/docs/en/cloud-paks/cp-data/3.5. 0?topic=services-watson-openscale

“Amazon sagemaker model monitor,” accessed 1 Sept 2022. [Online]. Available: https://docs.aws.amazon.com/sagemaker/latest/dg/ model-monitor.html

“Monitor azure machine learning,” accessed 1 Sept 2022. [Online]. Available: https://docs.microsoft.com/en-us/azure/ machine-learning/monitor-azure-machine-learning

“Vertex ai model monitoring,” accessed 1 Sept 2022. [Online]. Available: https://cloud.google.com/vertex-ai/docs/model-monitoring

W. H. Press and S. A. Teukolsky, “Kolmogorov-smirnov test for twodimensional data,” Computers in Physics, vol. 2, no. 74, 1988.

M. Greenwald and S. Khanna, “Space-efficient online computation of quantile summaries,” SIGMOD Rec., vol. 30, no. 2, p. 58–66, May 2001. [Online]. Available: https://doi.org/10.1145/376284.375670

T. Kailath, “The divergence and bhattacharyya distance measures in signal selection,” IEEE Trans. on Communication Technology, vol. 15, no. 1, pp. 52–60, 1967.

C. R. Harris, K. J. Millman, S. J. van der Walt, R. Gommers, P. Virtanen, D. Cournapeau, E. Wieser, J. Taylor, S. Berg, N. J. Smith, R. Kern, M. Picus, S. Hoyer, M. H. van Kerkwijk, M. Brett, A. Haldane, J. F. del R´ıo, M. Wiebe, P. Peterson, P. G´erard-Marchant, K. Sheppard, T. Reddy, W. Weckesser, H. Abbasi, C. Gohlke, and T. E. Oliphant, “Array programming with NumPy,” Nature, vol. 585, no. 7825, pp. 357–362, Sep. 2020. [Online]. Available: https://doi.org/10.1038/s41586-020-2649-2

P. Virtanen, R. Gommers, T. E. Oliphant, M. Haberland, T. Reddy, D. Cournapeau, E. Burovski, P. Peterson, W. Weckesser, J. Bright, S. J. van der Walt, M. Brett, J. Wilson, K. J. Millman, N. Mayorov, A. R. J. Nelson, E. Jones, R. Kern, E. Larson, C. J. Carey, ˙ I. Polat, Y. Feng, E. W. Moore, J. VanderPlas, D. Laxalde, J. Perktold, R. Cimrman, I. Henriksen, E. A. Quintero, C. R. Harris, A. M. Archibald, A. H. Ribeiro, F. Pedregosa, P. van Mulbregt, and SciPy 1.0 Contributors, “SciPy 1.0: Fundamental Algorithms for Scientific Computing in Python,” Nature Methods, vol. 17, pp. 261–272, 2020.

Wes McKinney, “Data Structures for Statistical Computing in Python,” in Proceedings of the 9th Python in Science Conference, St´efan van der Walt and Jarrod Millman, Eds., 2010, pp. 56– 61.

J. D. Hunter, “Matplotlib: A 2d graphics environment,” Computing in Science & Engineering, vol. 9, no. 3, pp. 90–95, 2007.

D. Vohra, Apache Parquet. Berkeley, CA: Apress, 2016, pp. 325–335. [Online]. Available: https://doi.org/10.1007/978-1-4842-2199-0

G. Vernik, M. Factor, E. K. Kolodner, P. Michiardi, E. Ofer, and F. Pace, “Stocator: Providing high performance and fault tolerance for apache spark over object storage,” in 2018 18th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGRID), 2018, pp. 462–471.