Abstract
Machine Learning (ML) has been taking significant evolutionary steps and provided sophisticated means in developing novel and smart, up-to-date applications. However, the development has also brought new types of hazards into the daylight that can have even destructive consequences required to be addressed. Evasion attacks are among the most utilized attacks that can be generated in adversarial settings during the system operation. In assumption, ML environment is benign, but in reality, perpetrators may exploit vulnerabilities to conduct these gradient-free or gradient-based malicious adversarial inference attacks towards cyber-physical systems (CPS), such as smart buildings. Evasion attacks provide a utility for perpetrators to modify, for example, a testing dataset of a victim ML-model. In this article, we conduct a literature review concerning evasion attacks and countermeasures and discuss how these attacks can be utilized in order to deceive the, i.e., CPS smart lock system’s ML-classifier to gain access to the smart building.
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
References
Alzantot M, Sharma Y, Chakraborty S, Zhang H, Hsieh C-J, Srivastava MB (2019) GenAttack: practical black-box attacks with gradient-free optimization (2019). arXiv:1805.11090v3 [cs.LG]
Carlini N, Wagner D (2017) Towards evaluating the robustness of neural networks. arXiv:1608.04644v2 [cs.CR]
Co KT (2018) Bayesian optimization for black-box evasion of machine learning systems. Imperial College London, Department of Computing
Coppola A, Stewardt BM (2014) LBFGS: efficient L-BFGS and OWL-QN optimization in R. http://cran.csiro.au/web/packages/lbfgs/vignettes/Vignette.pdf. Accessed 26 Aug 2020
De Groot J (2020) What is cyber security? Definition, best practices & more. Data Insider. https://digitalguardian.com/blog/what-cyber-security. Accessed 17 Aug 2020
DeepAI, What is defensive distillation? https://deepai.org/machine-learning-glossary-and-terms/defensive-distillation. Accessed 9 Oct 2019
Dewar RS (2014) The “Triptych of Cyber Security”: a classification of active cyber defence. In: 6th international conference on cyber conflict. NATO CCD COE Publications, Tallinn. https://www.ccdcoe.org/uploads/2018/10/d1r1s9_dewar.pdf. Accessed 17 Aug 2020
Gartner: Cybersecurity (2020). https://www.gartner.com/en/information-technology/glossary/cybersecurity. Accessed 2 Sept 2020
Goodfellow I, McDaniel P, Papernot N (2018) Making machine learning robust against adversarial inputs. Commun ACM 61(7):56–66
Ibitoye O, Abou-Khamis R, Matrawy A, Shafix MO (2019) The threat of adversarial attacks on machine learning in network security—a survey. arXiv:1911.02621v1 [cs.CR]
Jordan MI, Mitchell TM (2015) Machine learning: trends, perspectives, and prospects. Sciencemag.org. Science 349(6245)
LeCun Y, Bengio Y, Hinton G (2015) Deep learning. Nature 521(7553), 436–444
Lipton ZC, Berkowitz J, Elkan C (2015). A critical review of recurrent neural networks for sequence learning. arXiv preprint arXiv:1506.00019
Liu C, Ye D, Shang Y, Jiang S, Li S, Mei Y, Wang L (2020) Defend against adversarial samples by using perceptual hash. Comput Materials Continua (CMC) 62(3):1365–1386. https://doi.org/10.32605/cmc.2020.07421
Liu J, Zhang W, Yu N (2018) CAAD 2018: iterative ensemble adversarial attack. arXiv:1811.03456 [cs.CV]
Liu J, Zhang W, Zhang Y, Hou D, Liu Y, Zha H, Yu N (2018) Detection based defense against adversarial examples from the steganalysis point of view. arXiv:1806.09186v2 [cs.CV]
Loison A, Combey T, Hajri H (2020) Probabilistic Jacobian-based saliency maps attacks. arXiv:2007.06032 [cs.CV]
Luo Y, Boix X, Roig G, Poggio T, Zhao O (2015) Foveation-based mechanisms alleviate adversarial examples. arXiv:1511.06292
Ma S, Liu Y, Tao G, Lee WC, Zhang X (2019) NIC: detecting adversarial samples with neural network invariant checking. In: Network and distributed systems security (NDSS) symposium 2019, San Diego, CA, USA. https://doi.org/10.14722/ndss.2019.23415
Moisejevs I (2019) Evasion attacks on machine learning (or “adversarial examples”). In: Towards data science. https://towardsdatascience.com/evasion-attacks-on-machine-learning-or-adversarial-examples-12f2283e06a1. Accessed 30 July 2020
Moosavi-Dezfooli SM, Fawzi A, Fawzi O, Fossard P (2017) Universal adversarial perturbations. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 1765–1773
Okazaki N (2014) libLBFGS: a library of limited-memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS). http://www.chokkan.org/software/liblbfgs. Accessed 26 Aug 2020
Papernot N, McDaniel P, Goodfellow I, Jha S, Celik ZB, Swami A (2016) Practical black-box attacks against machine learning. arXiv:1602.02697 [cs.CR]
Pawlak A (2020) Adversarial attacks for fooling deep neural networks. https://neurosys.com/article/adversarial-attacks-for-fooling-deep-neural-networks. Accessed 31 July 2020
Ren K, Zheng T, Qin Z, Liu X (2020) Adversarial attacks and defences in deep learning. Engineering 6(3):346–360. https://doi.org/10.1016/j.eng.2019.12.012
Samangouei P, Kabhab M, Chellappa R (2018) Defense-GAN: protecting classifiers against adversarial attacks using generative models. arXiv:1805.06605v2 [cs.CV]
Sharif M, Bhagavatula S, Bauer L, Reiter MK (2016) Accessories to a crime: real and stealthy attacks on state-of-the-art face recognition. https://doi.org/10.1145/2976749.2978392
Short A, Pay TL, Gandhi A (2019) Defending against adversarial examples. In: Sandia report, SAND 2019-11748. Sandia National Laboratories
Song D (2019) Plenary session—toward trustworthy machine learning. In: Proceedings of a workshop of robust machine learning algorithms and systems for detection and mitigation of adversarial attacks and anomalies, pp 35–38
Szegedy C, Zaremba W, Sutskever I, Bruna J, Erhan D, Goodfellow I, Fergus R (2013) Intriguing properties of neural networks. arXiv:1312.6199
TensorflowCore: Adversarial Example Using FGSM (2020). https://www.tensorflow.org/tutorials/generative/adversarial_fgsm. Accessed 30 July 2020
Vorobeychik Y, Kantarcioglu M (2018) Adversarial machine learning synthesis lectures on artificial intelligence and machine learning. https://doi.org/10.2200/S00861ED1V01Y201806AIM039
Wiyatno R, Xu A (2018) Maximal Jacobian-based saliency map attack. arXiv:1808.07945v1 [cs.LG]
Wu H, Wang C, Tyshetskiy Y, Docherty A, Lu K, Zhu L (2019) Adversarial examples on graph data: deep insights into attack and defense. arXiv:1903.01610 [cs.LG]
Yanagita Y, Yamamura M (2018) Gradient masking is a type of overfitting. Int J Mach Learn Comput 8(3):203–207. https://doi.org/10.18178/ijmlc.2018.8.3.688
Yuan X, He P, Zhu Q, Bhat R, Li X (2017) Adversarial examples: attacks and defences for deep learning. arXiv:1712.07107 [cs.LG]
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Vähäkainu, P., Lehto, M., Kariluoto, A. (2021). Countering Adversarial Inference Evasion Attacks Towards ML-Based Smart Lock in Cyber-Physical System Context. In: Jahankhani, H., Jamal, A., Lawson, S. (eds) Cybersecurity, Privacy and Freedom Protection in the Connected World. Advanced Sciences and Technologies for Security Applications. Springer, Cham. https://doi.org/10.1007/978-3-030-68534-8_11
Download citation
DOI: https://doi.org/10.1007/978-3-030-68534-8_11
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-68533-1
Online ISBN: 978-3-030-68534-8
eBook Packages: Computer ScienceComputer Science (R0)