Development of a methodology for the protection of artificial intelligence systems in distributed information systems
https://doi.org/10.55648/1998-6920-2023-17-3-78-86
Abstract
This article is devoted to the methodology for the construction of artificial intelligent systems for information security tasks. It is planned to protect the artificial intelligence itself and the information that this device will process. The task is large-scale in its magnitude and the author of this text will try to present not just a concept, but rather an idea of how to achieve the goal of this result. By the result we will understand the final construction of a secure AI created for the organization of information security.
About the Author
S. I. Shterenberg
https://zss.sut.ru/shterenberg/
Moscow Technical University of Communications and Informatics (MTUCI)
Russian Federation
Moscow Technical University of Communications and Informatics (MTUCI)
Russian Federation
Stanislav I. Shterenberg - Cand. of Sci. (Engineering), Associate Professor of the Department of Information Security, MTUCI.
111024, Moscow, Aviamotornaya str., 8a
References
1. Bostrom N., Iskusstvennyi intellekt. Etapy. Ugrozy. Strategii [Artificial Intelligence. Stages. Threats. Strategies]. Moscow, Mann, Ivanov and Ferber, 2016. 496 p.
2. Shterenberg S. I. Metodika postroeniya zashchishchennykh sistem iskusstvennogo intellekta dlya provedeniya elektroretinografii v oftal'mologii [Methods of constructing protected artificial intelligence systems for conducting electroretinography in ophthalmology]. Oftal'mokhirurgiya, 2022, no. S4. pp. 51-57.
3. Nikolenko S., Kadurin A., Arkhangel'skaya E. Glubokoe obuchenie [Deep learning]. Saint Petersburg, Peter, 2018. 480 p.
4. Ushakov I. A. Obnaruzhenie insaiderovv korporativnoi komp'yuternoi seti na osnove tekhnologii analiza bol'shikh dannykh [Detection of insiders in the corporate computer network based on big data analysis technologies]. Vestnik Sankt-Peterburgskogo gosudarstvenno-go universiteta tekhnologii i dizaina. Seriya 1: Estestvennye i tekhnicheskie nauki, 2019, no. 4. pp. 38-43.
5. Stepanov M. D., Pavlenko E. Yu., Lavrova D. S. Obnaruzhenie setevykh atak v programmnokonfiguriruemykh setyakh s ispol'zovaniem algoritma izoliruyushchego lesa [Detection of network at- tacks in software-configurable networks using the isolating forest algorithm]. Problemy informatsionnoi bezopasnosti. Komp'yuternye sistemy, 2021, no. 1, pp. 62-78.
6. Vasil'eva K. V., Lavrova D. S. Obnaruzhenie anomalii v kiberfizicheskikh sistemakh s ispol'zovaniem grafovykh neironnykh setei [Detection of anomalies in cyberphysical systems using graph neural networks]. Problemy informatsionnoi bezopasnosti. Komp'yuternye sistemy, 2021, no. 1. pp. 117-130.
7. Khapke Kh., Nel'son K. Razrabotka konveierov mashinnogo obucheniya. Avtomatizatsiya zhiz-nennykh tsiklov modeli s pomoshch'yu TensorFlow [Development of machine learning pipelines. Automation of life cycles of the model using TensorFlow]. Moscow, DMK Press, 2021. 346 p.
8. Minyaev A.A. Metod i metodika otsenki effektivnosti sistemy zashchity territorial'no-raspredelennykh informatsionnykh sistem [Method and methodology for evaluating the effectiveness of the protection system of territorial-distributed information systems]. Informatizatsiya i svyaz', 2020, no. 6, pp. 29-36.
9. Gamidov T.O., Vitkova L.A., Kovtsur M.M. Razrabotka modelei i algoritmov analiza dannykh dlya issledovaniya khoda intsidentov i krizisov v sotsial'nykh setyakh [Development of models and algorithms for data analysis to study the course of incidents and crises in social networks]. Vestnik SanktPeterburgskogo gosudarstvennogo universiteta tekhnologii i dizaina. Seriya 1: Estestvennye i tekhnicheskie nauki, 2020, no. 2, pp. 3-10.
Supplementary files
Review
For citations:
Shterenberg S.I. Development of a methodology for the protection of artificial intelligence systems in distributed information systems. The Herald of the Siberian State University of Telecommunications and Information Science. 2023;17(3):78-86. (In Russ.) https://doi.org/10.55648/1998-6920-2023-17-3-78-86
Views:
305
Email this article 
