Abstract
In this paper, we present a methodology for constructing explanations for AI classification algorithms. The methodology consists of constructing a model of the context of the application in the Isabelle Infrastructure framework (IIIf) and an algorithm that allows to extract a precise logical rule that specifies the behaviour of the black box algorithm thus allowing to explain it. The explanation is given within the rich logical model of the IIIf. It is thus suitable for human audiences. We illustrate this and validate the methodology on the application example of credit card scoring with special relation to the right of explanation as given by the GDPR.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The latter two type definitions are omitted for brevity.
- 2.
Note, that the interleaving of the CTL-operators AG and EF with logical operators, like implication \(\longrightarrow \) is only possible since we use a Higher Order logic embedding of CTL.
- 3.
References
Arrieta, A.B., et al.: Explainable artificial intelligence (XAI): concepts, taxonomies, opportunities and challenges toward responsible AI. Inf. Fusion 58, 82–115 (2020)
Belle, V., Papantonis, I.: Principles and practice of explainable machine learning. CoRR, abs/2009.11698 (2020)
Bull, S.: London boroughs mapped and ranked by residents’ credit scores - how money-savvy is your area? Accessed 22 July 2022
CHIST-ERA. Success: Secure accessibility for the internet of things (2016). http://www.chistera.eu/projects/success
Gilpin, L.H., Bau, D., Yuan, B.Z., Bajwa, A., Specter, M.A., Kagal, L.: Explaining explanations: an approach to evaluating interpretability of machine learning. CoRR, abs/1806.00069 (2018)
T. is Money. How well do your neighbours manage their money? Interactive map reveals average credit scores by postcode. https://www.thisismoney.co.uk/money/cardsloans/article-3273996/How-neighbours-manage-money-Interactive-map-reveals-average-credit-scores-postcode.html. Accessed 22 July 2022
Kammüller, F.: A proof calculus for attack trees in Isabelle. In: Garcia-Alfaro, J., Navarro-Arribas, G., Hartenstein, H., Herrera-Joancomartí, J. (eds.) ESORICS/DPM/CBT 2017. LNCS, vol. 10436, pp. 3–18. Springer, Cham (2017). https://doi.org/10.1007/978-3-319-67816-0_1
Kammüller, F.: Attack trees in Isabelle. In: Naccache, D., et al. (eds.) ICICS 2018. LNCS, vol. 11149, pp. 611–628. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01950-1_36
Kammüller, F.: Formal modeling and analysis of data protection for GDPR compliance of IoT healthcare systems. In: IEEE Systems, Man and Cybernetics, SMC 2018. IEEE (2018)
Kammüller, F.: Attack trees in Isabelle extended with probabilities for quantum cryptography. Comput. Secur. 87 (2019)
Kammüller, F.: Combining secure system design with risk assessment for IoT healthcare systems. In: Workshop on Security, Privacy, and Trust in the IoT, SPTIoT 2019, co-located with IEEE PerCom. IEEE (2019)
Kammüller, F.: Dependability engineering in Isabelle (2021). arXiv preprint, http://arxiv.org/abs/2112.04374
Kammüller, F.: Explanation by automated reasoning using the Isabelle infrastructure framework (2021). arXiv preprint, http://arxiv.org/abs/2112.14809
Kammüller, F.: Isabelle Insider and Infrastructure framework with Kripke strutures, CTL, attack trees, security refinement, and examples including IoT, GDPR, QKD, social networks, and credit scoring (2022). https://github.com/flokam/IsabelleAT
Kammüller, F., Alvarado, C.M.: Exploring rationality of self awareness in social networking for logical modeling of unintentional insiders (2021). arXiv preprint, http://arxiv.org/abs/2111.15425
Kammüller, F., Kerber, M.: Applying the Isabelle insider framework to airplane security. Sci. Comput. Program. 206 (2021)
Kammüller, F., Kerber, M., Probst, C.: Insider threats for auctions: Formal modeling, proof, and certified code. J. Wirel. Mob. Netw. Ubiquit. Comput. Dependable Appl. (JoWUA) 8(1) (2017)
Kammüller, F., Lutz, B.: Modeling and analyzing the corona-virus warning app with the Isabelle infrastructure framework. In: Garcia-Alfaro, J., Navarro-Arribas, G., Herrera-Joancomarti, J. (eds.) DPM/CBT 2020. LNCS, vol. 12484, pp. 128–144. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-66172-4_8
Kammüller, F., Nestmann, U.: Inter-blockchain protocols with the Isabelle infrastructure framework. In: Formal Methods for Blockchain, 2nd International Workshop, co-located with CAV 2020, Open Access Series in Informatics. Dagstuhl Publishing (2020, to appear)
Kammüller, F., Wenzel, M., Paulson, L.C.: Locales a sectioning concept for Isabelle. In: Bertot, Y., Dowek, G., Théry, L., Hirschowitz, A., Paulin, C. (eds.) TPHOLs 1999. LNCS, vol. 1690, pp. 149–165. Springer, Heidelberg (1999). https://doi.org/10.1007/3-540-48256-3_11
Myers, A.C., Liskov, B.: Complete, safe information flow with decentralized labels. In: Proceedings of the IEEE Symposium on Security and Privacy. IEEE (1999)
Nipkow, T., Wenzel, M., Paulson, L.C. (eds.): Isabelle/HOL – A Proof Assistant for Higher-Order Logic. LNCS, vol. 2283. Springer, Heidelberg (2002). https://doi.org/10.1007/3-540-45949-9
Pieters, W.: Explanation and trust: what to tell the user in security and AI? Ethics Inf. Technol. 13(1), 53–64 (2011)
Viganó, L., Magazzeni, D.: Explainable security. In: IEEE European Symposium on Security and Privacy Workshops, EuroS &PW. IEEE (2020)
Wachter, S., Mittelstadt, B., Russell, C.: Counterfactual explanations without opening the black box: automated decisions and the GDPR. Harvard J. Law Technol. 31(2) (2018)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Kammüller, F. (2023). Explanation of Black Box AI for GDPR Related Privacy Using Isabelle. In: Garcia-Alfaro, J., Navarro-Arribas, G., Dragoni, N. (eds) Data Privacy Management, Cryptocurrencies and Blockchain Technology. DPM CBT 2022 2022. Lecture Notes in Computer Science, vol 13619. Springer, Cham. https://doi.org/10.1007/978-3-031-25734-6_5
Download citation
DOI: https://doi.org/10.1007/978-3-031-25734-6_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-25733-9
Online ISBN: 978-3-031-25734-6
eBook Packages: Computer ScienceComputer Science (R0)