Abstract
Understanding predictions made by Machine Learning models is critical in many applications. In this work, we investigate the performance of two methods for explaining tree-based models: ‘Tree Interpreter (TI)’ and ‘SHapley Additive exPlanations TreeExplainer (SHAP-TE)’. Using a case study on detecting anomalies in job runtimes of applications that utilize cloud-computing platforms, we compare these approaches using a variety of metrics, including computation time, significance of attribution value, and explanation accuracy. We find that, although the SHAP-TE offers consistency guarantees over TI, at the cost of increased computation, consistency does not necessarily improve the explanation performance in our case study.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
Feature Attribution (FA) is defined as the contribution each independent variable or a “feature” made to the final prediction of a model.
- 2.
See Sect. 2 for the definition of consistency.
- 3.
Feature Attribution Method (FAM), referred to as the explanation method that calculates FAs to interpret each prediction generated by a model.
- 4.
Some of the covariate variables in postgreSQL dataset are continuous, which when grouped reduces the number of data points per cluster.
- 5.
RBO implementation: https://github.com/changyaochen/rbo.
- 6.
Dataset can be found at https://groups.cs.umass.edu/kdl/causal-eval-data.
- 7.
This data is collected in the work by [5].
- 8.
For eg, consider 2 lists of attribution values \(S_1=[1, 1.1, 1.3]\) and \(S_2=[1, 3, 5]\). The ranking obtained from values in \(S_2\) is more reliable than \(S_1\).
- 9.
References
Caruana, R., Karampatziakis, N., Yessenalina, A.: An empirical evaluation of supervised learning in high dimensions. In: Proceedings of the 25th International Conference on Machine Learning, ICML 2008, pp. 96–103 (2008)
Chandola, V., Banerjee, A., Kumar, V.: Anomaly detection: a survey. ACM Comput. Surv. 41, 1–58 (2009)
Cuzzocrea, A., Mumolo, E., Cecolin, R.: Runtime anomaly detection in embedded systems by binary tracing and hidden Markov models. In 2015 IEEE 39th Annual Computer Software and Applications Conference, vol. 2, pp. 15–22 (2015)
Duque Anton, S., Sinha, S., Schotten, H.: Anomaly-based intrusion detection in industrial data with SVM and random forests, pp. 1–6 (2019)
Gentzel, A., Garant, D., Jensen, D.: The case for evaluating causal models using interventional measures and empirical data. In: Wallach, H., Larochelle, H., Beygelzimer, A., d’Alché-Buc, F., Fox, E., Garnett, R. (eds.) Advances in Neural Information Processing Systems, vol. 32, pp. 11722–11732. Curran Associates Inc. (2019)
Kuhn, H.W., Tucker, A.W.: Contributions to the Theory of Games, vol. 2. Princeton University Press, Princeton (1953)
Lipovetsky, S., Conklin, M.: Analysis of regression in game theory approach. Appl. Stochast. Models Bus. Ind. 17, 319–330 (2001)
Lundberg, S.M., et al.: From local explanations to global understanding with explainable AI for trees. Nat. Mach. Intell. 2(1), 2522–5839 (2020)
Lundberg, S.M., Lee, S.-I.: A unified approach to interpreting model predictions. In: Advances in Neural Information Processing Systems, vol. 30 (2017)
Peiris, M., Hill, J.H., Thelin, J., Bykov, S., Kliot, G., Konig, C.: PAD: performance anomaly detection in multi-server distributed systems. In: 2014 IEEE 7th International Conference on Cloud Computing, pp. 769–776 (2014)
Primartha, R., Tama, B.A.: Anomaly detection using random forest: a performance revisited. In: 2017 International Conference on Data and Software Engineering (ICoDSE), pp. 1–6 (2017)
Ribeiro, M.T., Singh, S., Guestrin, C.: “why should I trust you?”: explaining the predictions of any classifier. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA, 13–17 August 2016, pp. 1135–1144 (2016)
Saabas, A.: Treeinterpreter. https://github.com/andosa/treeinterpreter
Shao, L., et al.: Griffon. In: Proceedings of the ACM Symposium on Cloud Computing - SoCC 2019 (2019)
Shrikumar, A., Greenside, P., Kundaje, A.: Learning important features through propagating activation differences. CoRR abs/1704.02685 (2017)
Shrikumar, A., Greenside, P., Shcherbina, A., Kundaje, A.: Not just a black box: learning important features through propagating activation differences. CoRR abs/1605.01713 (2016)
Štrumbelj, E., Kononenko, I.: Explaining prediction models and individual predictions with feature contributions. Knowl. Inf. Syst. 41, 647–665 (2013)
Sultani, W., Chen, C., Shah, M.: Real-world anomaly detection in surveillance videos. In: The IEEE Conference on Computer Vision and Pattern Recognition (2018)
Webber, W., Moffat, A., Zobel, J.: A similarity measure for indefinite rankings. ACM Trans. Inf. Syst. 28, 4 (2010)
Wulsin, D., Blanco, J., Mani, R., Litt, B.: Semi-supervised anomaly detection for EEG waveforms using deep belief nets. In: 2010 Ninth International Conference on Machine Learning and Applications, pp. 436–441 (2010)
Acknowledgements
We thank our mentors, Javier Burroni and Prof. Andrew McCallum, for their guidance. We also thank Minsoo Thigpen for organizational support, as well as Scott Lundberg for providing insightful suggestions on a earlier draft. Finally, we thank anonymous reviewers for their feedback.
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Sharma, P. et al. (2020). Evaluating Tree Explanation Methods for Anomaly Reasoning: A Case Study of SHAP TreeExplainer and TreeInterpreter. In: Grossmann, G., Ram, S. (eds) Advances in Conceptual Modeling. ER 2020. Lecture Notes in Computer Science(), vol 12584. Springer, Cham. https://doi.org/10.1007/978-3-030-65847-2_4
Download citation
DOI: https://doi.org/10.1007/978-3-030-65847-2_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-65846-5
Online ISBN: 978-3-030-65847-2
eBook Packages: Computer ScienceComputer Science (R0)