Abstract
Today’s world is witnessing a shift from human-written software to machine-learned software, with the rise of systems that rely on machine learning. These systems typically operate in non-static environments, which are prone to unexpected changes, as is the case of self-driving cars and enterprise systems. In this context, machine-learned software can misbehave. Thus, it is paramount that these systems are capable of detecting problems with their machined-learned components and adapting themselves to maintain desired qualities. For instance, a fraud detection system that cannot adapt its machine-learned model to efficiently cope with emerging fraud patterns or changes in the volume of transactions is subject to losses of millions of dollars. In this paper, we take a first step towards the development of a framework for self-adaptation of systems that rely on machine-learned components. We describe: (i) a set of causes of machine-learned component misbehavior and a set of adaptation tactics inspired by the literature on machine learning, motivating them with the aid of two running examples from the enterprise systems and cyber-physical systems domains; (ii) the required changes to the MAPE-K loop, a popular control loop for self-adaptive systems; and (iii) the challenges associated with developing this framework. We conclude with a set of research questions to guide future work.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
In ML, data cleaning corresponds to the process of identifying and correcting errors in a dataset that may negatively impact a predictive model.
- 2.
Fraud detection systems normally rely on a fixed set of humans at any given time. This determines a maximum load of transactions that can be processed with a human in the loop.
References
Abedjan, Z., et al.: Detecting data errors: where are we and what needs to be done? Proc. VLDB 9(12), 19993–1004 (2016)
Alipourfard, O., et al.: CherryPick: adaptively unearthing the best cloud configurations for big data analytics. In: Proceedings of NSDI (2017)
Aparício, D., et al.: Arms: automated rules management system for fraud detection. arXiv preprint arXiv:2002.06075 (2020)
Badue, C., Guidolini, R., et al.: Self-driving cars: a survey. Expert Syst. App. 165, 113816 (2021)
Bartocci, E., et al.: Specification-based monitoring of cyber-physical systems: a survey on theory, tools and applications. In: Bartocci, E., Falcone, Y. (eds.) Lectures on Runtime Verification. LNCS, vol. 10457, pp. 135–175. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-75632-5_5
Bojarski, M., et al.: End to end learning for self-driving cars. arXiv preprint arXiv:1604.07316 (2016)
Bouchabou, D., Nguyen, S.M., Lohr, C., LeDuc, B., Kanellos, I.: A survey of human activity recognition in smart homes based on iot sensors algorithms: taxonomies, challenges, and opportunities with deep learning. Sensors 21(18), 6037 (2021)
Branco, B., et al.: Interleaved sequence RNNs for fraud detection. In: Proceedings of KDD (2020)
Breiman, L.: Bagging predictors. Mach. Learn. 24, 123–140 (1996). https://doi.org/10.1007/BF00058655
Bureš, T.: Self-adaptation 2.0. In: 2021 International Symposium on Software Engineering for Adaptive and Self-Managing Systems (SEAMS) (2021)
Cámara, J., Lopes, A., Garlan, D., Schmerl, B.: Adaptation impact and environment models for architecture-based self-adaptive systems. Sci. Comput. Program. 127, 50–75 (2016)
Cao, Y., Yang, J.: Towards making systems forget with machine unlearning. In: Proceedings of S &P. IEEE (2015)
Cao, Y., et al.: Efficient repair of polluted machine learning systems via causal unlearning. In: Proceedings of Asia CCS (2018)
Casimiro, M., Romano, P., Garlan, D., Moreno, G., Kang, E., Klein, M.: Self-adaptation for machine learning based systems. In: Proceedings of SAML (2021)
Casimiro, M., Garlan, D., Cámara, J., Rodrigues, L., Romano, P.: A probabilistic model checking approach to self-adapting machine learning systems. In: Procseedings of ASYDE, Co-located with SEFM 2021 (2021)
Casimiro, M., et al.: Lynceus: cost-efficient tuning and provisioning of data analytic jobs. In: Proceedings of ICDCS (2020)
Chen, T.: All versus one: an empirical comparison on retrained and incremental machine learning for modeling performance of adaptable software. In: Proceedings of SEAMS. IEEE (2019)
Chen, Z., Huang, X.: End-to-end learning for lane keeping of self-driving cars. In: Proceedings of IV (2017)
Cheng, B.H.C., et al.: Software engineering for self-adaptive systems: a research roadmap. In: Cheng, B.H.C., de Lemos, R., Giese, H., Inverardi, P., Magee, J. (eds.) Software Engineering for Self-Adaptive Systems. LNCS, vol. 5525, pp. 1–26. Springer, Heidelberg (2009). https://doi.org/10.1007/978-3-642-02161-9_1
Cheng, H.T., et al.: Wide & deep learning for recommender systems. In: Proceedings of DLRS (2016)
Cheng, S.W., et al.: Evaluating the effectiveness of the rainbow self-adaptive system. In: Proceedings of SEAMS. IEEE (2009)
Christi, A., et al.: Evaluating fault localization for resource adaptation via test-based software modification. In: Proceedings of QRS (2019)
Cito, J., Dillig, I., Kim, S., Murali, V., Chandra, S.: Explaining mispredictions of machine learning models using rule induction. In: Proceedings of ESEC/FSE (2021)
Cruz, A.F., et al.: A bandit-based algorithm for fairness-aware hyperparameter optimization. CoRR abs/2010.03665 (2020)
deGrandis, P., Valetto, G.: Elicitation and utilization of application-level utility functions. In: Proceedings of ICAC (2009)
Elsken, T., Metzen, J.H., Hutter, F.: Neural architecture search: a survey. J. Mach. Learn. Res. 20(1), 1997–2017 (2019)
Erickson, B.J., et al.: Machine learning for medical imaging. Radiographics 37(2), 505 (2017)
Esrafilian-Najafabadi, M., Haghighat, F.: Occupancy-based HVAC control systems in buildings: a state-of-the-art review. Build. Environ. 197, 107810 (2021)
Gao, D., Liu, Y., Huang, A., Ju, C., Yu, H., Yang, Q.: Privacy-preserving heterogeneous federated transfer learning. In: 2019 IEEE International Conference on Big Data (Big Data), pp. 2552–2559. IEEE (2019)
Ghahremani, S., Giese, H., Vogel, T.: Improving scalability and reward of utility-driven self-healing for large dynamic architectures. ACM Trans. Auton. Adapt. Syst. 14(3), 1–41 (2020)
Gheibi, O., Weyns, D.: Lifelong self-adaptation: self-adaptation meets lifelong machine learning. In: Proceedings of SEAMS (2022)
Gheibi, O., et al.: Applying machine learning in self-adaptive systems: a systematic literature review. arXiv preprint arXiv:2103.04112 (2021)
Gu, T., et al.: BadNets: evaluating backdooring attacks on deep neural networks. IEEE Access 7, 47230–47244 (2019)
Guo, X., Shen, Z., Zhang, Y., Wu, T.: Review on the application of artificial intelligence in smart homes. Smart Cities 2(3), 402–420 (2019)
Huang, L., et al.: Adversarial machine learning. In: Proceedings of AISec (2011)
Huchuk, B., Sanner, S., O’Brien, W.: Comparison of machine learning models for occupancy prediction in residential buildings using connected thermostat data. Build. Environ. 160, 106177 (2019)
Jamshidi, P., et al.: Machine learning meets quantitative planning: enabling self-adaptation in autonomous robots. In: Proceedings of SEAMS (2019)
Kephart, J.O., Chess, D.M.: The vision of autonomic computing. Computer 36(1), 46–50 (2003)
Krupitzer, C., et al.: A survey on engineering approaches for self-adaptive systems. Pervasive Mob. Comput. 17, 184–206 (2018)
Langford, M.A., Chan, K.H., Fleck, J.E., McKinley, P.K., Cheng, B.H.: MoDALAS: model-driven assurance for learning-enabled autonomous systems. In: Proceedings of MODELS (2021)
Liu, B.: Learning on the job: online lifelong and continual learning. In: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 34 (2020)
Liu, Y., et al.: A secure federated transfer learning framework. Proc. IS 35(4), 70–82 (2020)
Lucas, Y., Jurgovsky, J.: Credit card fraud detection using machine learning: a survey. CoRR abs/2010.06479 (2020)
Mallozzi, P., Pelliccione, P., Knauss, A., Berger, C., Mohammadiha, N.: Autonomous vehicles: state of the art, future trends, and challenges. In: Dajsuren, Y., van den Brand, M. (eds.)Automotive Systems and Software Engineering, pp. 347–367. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-12157-0_16
Mendes, P., et al.: TrimTuner: Efficient optimization of machine learning jobs in the cloud via sub-sampling. In: MASCOTS (2020)
Miller, B., et al.: Reviewer integration and performance measurement for malware detection. In: Proceedings of DIMVA (2016)
Moreno, G.A., et al.: Flexible and efficient decision-making for proactive latency-aware self-adaptation. ACM Trans. Auton. Adapt. Syst. 13(1), 1–36 (2018)
Moreno-Torres, J.G., Raeder, T., Alaiz-Rodríguez, R., Chawla, N.V., Herrera, F.: A unifying view on dataset shift in classification. Pattern Recogn. 45(1), 521–530 (2012)
Nguyen, C., Hassner, T., Seeger, M., Archambeau, C.: LEEP: a new measure to evaluate transferability of learned representations. In: Proceedings of ICML. PMLR (2020)
Osborne, M.A., et al.: Gaussian processes for global optimization. In: LION (2009)
Ovadia, Y., et al.: Can you trust your model’s uncertainty? evaluating predictive uncertainty under dataset shift. In: Proceedings of NIPS (2019)
Pan, S.J., Yang, Q.: A survey on transfer learning. IEEE TKDE 22(10), 1345–4350 (2009)
Pandey, A., Moreno, G.A., Cámara, J., Garlan, D.: Hybrid planning for decision making in self-adaptive systems. In: Proceedings of SASO (2016)
Papamartzivanos, D., et al.: Introducing deep learning self-adaptive misuse network intrusion detection systems. IEEE Access 7, 13546–13560 (2019)
Peng, Z., Yang, J., Chen, T.H., Ma, L.: A first look at the integration of machine learning models in complex autonomous driving systems: a case study on Apollo. In: Proceedings of ESEC/FSE (2020)
Pinto, F., et al.: Automatic model monitoring for data streams. arXiv preprint arXiv:1908.04240 (2019)
Quionero-Candela, J., et al.: Dataset Shift in Machine Learning. The MIT Press, Cambridge (2009)
Rabanser, S., et al.: Failing loudly: an empirical study of methods for detecting dataset shift. In: Proceedings of NIPS (2019)
Saputri, T.R.D., Lee, S.W.: The application of machine learning in self-adaptive systems: a systematic literature review. IEEE Access 8, 205948–205967 (2020)
Shi, J., Yu, N., Yao, W.: Energy efficient building HVAC control algorithm with real-time occupancy prediction. Energy Proc. 111, 267–276 (2017)
Silver, D.L., Yang, Q., Li, L.: Lifelong machine learning systems: beyond learning algorithms. In: 2013 AAAI Spring Symposium Series (2013)
Singh, A., Sikdar, B.: Adversarial attack for deep learning based IoT appliance classification techniques. In: 2021 IEEE 7th World Forum on Internet of Things (WF-IoT). IEEE (2021)
Surantha, N., Wicaksono, W.R.: Design of smart home security system using object recognition and PIR sensor. Proc. Comput. Sci. 135, 465–472 (2018)
Swersky, K., et al.: Multi-task Bayesian optimization. Proc. NIPS 26, 1–9 (2013)
Wang, Z.J., Choi, D., Xu, S., Yang, D.: Putting humans in the natural language processing loop: a survey. arXiv preprint arXiv:2103.04044 (2021)
Wu, D., et al.: A highly accurate framework for self-labeled semisupervised classification in industrial applications. IEEE TII 14(3), 1–12 (2018)
Wu, Y., et al.: DeltaGrad: rapid retraining of machine learning models. In: Proceedings of ICML (2020)
Xiao, Y., et al.: Self-checking deep neural networks in deployment. In: Proceedings of ICSE (2021)
Yadwadkar, N.J., Hariharan, B., Gonzalez, J.E., Smith, B., Katz, R.H.: Selecting the \(<\)i\(>\)best\(<\)/i\(>\) vm across multiple public clouds: a data-driven performance modeling approach. In: Proceedings of SoCC, pp. 452–465 (2017)
Yang, J., Zhou, K., Li, Y., Liu, Z.: Generalized out-of-distribution detection: a survey. arXiv preprint arXiv:2110.11334 (2021)
Yang, Z., Asyrofi, M.H., Lo, D.: BiasRV: uncovering biased sentiment predictions at runtime. CoRR abs/2105.14874 (2021)
Zhou, X., Lo Faro, W., Zhang, X., Arvapally, R.S.: A framework to monitor machine learning systems using concept drift detection. In: Abramowicz, W., Corchuelo, R. (eds.) BIS 2019. A Framework to Monitor Machine Learning Systems Using Concept Drift Detection, vol. 353, pp. 218–231. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-20485-3_17
Acknowledgements
Support for this research was provided by Fundação para a Ciência e a Tecnologia (Portuguese Foundation for Science and Technology) through the Carnegie Mellon Portugal Program under Grant SFRH/BD/150643/2020 and via projects with references POCI-01-0247-FEDER-045915, POCI-01-0247-FEDER-045907, and UIDB/50021/2020. The contributions of Gabriel Moreno and Mark Klein are based upon work funded and supported by the Department of Defense under Contract No. FA8702-15-D-0002 with Carnegie Mellon University for the operation of the Software Engineering Institute, a federally funded research and development center. Such contributions are used with permission, but ownership of the underlying intellectual property embodied within such contributions is retained by Carnegie Mellon University. DM22-0149.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Casimiro, M., Romano, P., Garlan, D., Moreno, G.A., Kang, E., Klein, M. (2022). Self-adaptive Machine Learning Systems: Research Challenges and Opportunities. In: Scandurra, P., Galster, M., Mirandola, R., Weyns, D. (eds) Software Architecture. ECSA 2021. Lecture Notes in Computer Science, vol 13365. Springer, Cham. https://doi.org/10.1007/978-3-031-15116-3_7
Download citation
DOI: https://doi.org/10.1007/978-3-031-15116-3_7
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-15115-6
Online ISBN: 978-3-031-15116-3
eBook Packages: Computer ScienceComputer Science (R0)