Machine Learning Drift: spatial and temploral analysis

Machine Learning models evolve due to evolution of the input data, the relationtions between predictors and dependent variables or even due to temporal degration of the implementation.

The implications of these phenomenons for machine learning operations requires not just the proper identification, but also the understanding and the adaptation of the systems, adapting risk models, security frameworks and ultimately the software supply chain.

This presentation review the state of the art on these topics, presenting the result of recent research papers and highlighting the elements still to be defined.

Transcript

1. ML Model Drifting Spatial & Temporal Analysis Foto de Mike Hindle en Unsplash Andrés L. Martínez Ortiz @davilagrau
2. About me… Andrés-Leonardo Martínez-Ortiz a.k.a almo, holds a PhD on Software, Systems and Computing and a Master on Computer Science. Based on Zurich, almo is a member of the Google Machine Learning Site Reliability Engineering team, leading several programs aiming for reliability, efficiency & convergence. He is also a member of IEEE, ACM, Linux Foundation and Computer Society. @davilagrau almo
3. Agenda Machine Learning Operations: • Efficiency, Reliability and convergence. • Model Drift Spatial Drift Temporal Drift Risk Modelling References Photo by Bradyn Shock on Unsplash
4. Machine Learning Operations Photo by Philipp Katzenberger on Unsplash
5. Machine Learning Operations
6. MLOps: Model Drifting Machine Learning Abstract Model Data Drifting Concept Drifting Temporal Drifting Spatial Drifting Temporal Drifting
7. Spatial Drifting Photo by Pawel Czerwinski on Unsplash
8. Spatial Drift: Challenges and research areas • Detection under unstructured and noise datasets • Understanding of the model drift is required for a proper treatment. • Reacting to model drift, adapting the life cycle. Photo by Harole Ethan on Unsplash
9. Detection General Framework Source: arXiv:2004.05785 Inferring data distribution Extracting sensitive features Severity of the drift Drift detection Accuracy
10. General drift partners and algorithms’ performance Source: arXiv:2004.05785 Source: Expert Systems with Applications 41 (2014) 8144–8156
11. Drift Understanding Time series analysis Synthetic data Degradation patterns datasets Explainable analysis (symbolic regression) For critical applications, detection is not enough. ML drift presents high dependency on the application, making difficult general solutions. Open dataset, synthetic data are opportunities for new developments. Massive Online Analysis (MOA)
12. Temporal Drift Photo by Jon Tyson on Unsplash
13. What is the temporal ml drifting? Temporal degradation of ml models affecting • Penalized Regression • Random Forest • Gradient Boosting • Neural network over • long life datasets (3-5 years) • with no data or concept drifting • Multi-domain: weather, financial, hospital planning and flight delays. Foto de Dustin Humes en Unsplash Vela, D., Sharp, A., Zhang, R. et al. Temporal quality degradation in AI models. Sci Rep 12, 11654 (2022).
14. Evaluation framework Vela, D., Sharp, A., Zhang, R. et al. Temporal quality degradation in AI models. Sci Rep 12, 11654 (2022). You need to define your own evaluation framework.
15. How does the temporal ML drifting look like? No degradation or gradual Vela, D., Sharp, A., Zhang, R. et al. Temporal quality degradation in AI models. Sci Rep 12, 11654 (2022).
16. How does the temporal ML drifting look like? Explosive degradation Vela, D., Sharp, A., Zhang, R. et al. Temporal quality degradation in AI models. Sci Rep 12, 11654 (2022).
17. How does the temporal ML drifting look like? Increasing variability Vela, D., Sharp, A., Zhang, R. et al. Temporal quality degradation in AI models. Sci Rep 12, 11654 (2022).
18. How does the temporal ML drifting look like? Exotic patterns: chaos and periodic Vela, D., Sharp, A., Zhang, R. et al. Temporal quality degradation in AI models. Sci Rep 12, 11654 (2022).
19. Implications for ML Operations Long lasting models demands temporal degradation analysis. Numerical analysis and dynamic systems analysis are required. Automatic re-training is not (always) an option: • Lack of clear thresholds • Lack of training data • Lack of convergence • Catastrophic forgetting • Random seeds dependencies Recommendations Extend drift analysis, including temporal drift. Evaluation should include models, hyperparameters and size of training data. Feature drifting analysis Real time or high frequency monitoring Your models life is now longer than ever. Temporal drift analysis is a must.
20. Risk Modelling and Mitigation Photo by Edge2Edge Media on Unsplash
21. Risks Vulnerabilities in complex systems (hardwares, software, data) Deceived Human Interaction: misleading reporting Unpredictable sequential planning (collusion) Difficulties being shut down Photo by Tom Morel on Unsplash Deployment of ML systems required risk analysis, including technological, business and legal perspectives
22. Model Evaluation along the life cycle Internal: multi-layer APIs: red teaming Auditing Evaluation requires proper development, documentation and deployment, adding extra complexity External
23. Model Evaluation Source: arXiv:2305.15324
24. Limitations and hazards Limitations Complex system integrations: unpredictable interactions Unknown unknown Hidden Features Over-promising Evaluation technology Hazards Impact of model evaluation Superficial improvements to model safety
25. Organizational Implications Communications • Incident analysis and reporting, including external parties. • Auditability • Scientific peer-review • Internal communication for business units and non technical staff. Security • Intensive evaluation strategies • AI-based monitoring • Fast responses protocols • Integrity verification, authorization and auditing Photo by Alexander Grey on Unsplash
26. Thank you! Questions?
27. References • Lu J., Liu A., Dong F., Gu F., Gama J. and Zhang G. Learning under Concept Drift: A Review, arXiv:2004.05785 (2020). (link) • Zeniseka, J., Holzingera, F. and Affenzellera, M. Machine learning based concept drift detection for predictive maintenance, Computers & Industrial Engineering 137 (2019) 106031. • Gonçalves P. M., Carvalho Santos S.G.T., Barros, R.S.M. and Vieira D.C.L. A comparative study on concept drift detectors, Expert Systems with Applications 41 (2014) 8144–8156 • Vela, D., Sharp, A., Zhang, R. et al. Temporal quality degradation in AI models. Sci Rep 12, 11654 (2022). (link) • Shevlane T., Farquhar S., Garfinkel B., Phuong M., Whittlestone J., Leung J, Kokotajlo D., Marchal N., Anderljung M., Kolt N., Ho L., Siddarth D., Avin S., Hawkins W., Kim B., Gabriel I., Bolina V., Clark J., Bengio Y., Christiano P. and Dafoe A. Model evaluation for extreme risks, arXiv:2305.15324 (link)