<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Publishing DTD v1.3 20210610//EN" "JATS-journalpublishing1-3.dtd">
<article article-type="research-article" dtd-version="1.3" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xml:lang="ru"><front><journal-meta><journal-id journal-id-type="publisher-id">ntv</journal-id><journal-title-group><journal-title xml:lang="ru">Научно-технический вестник информационных технологий, механики и оптики</journal-title><trans-title-group xml:lang="en"><trans-title>Scientific and Technical Journal of Information Technologies, Mechanics and Optics</trans-title></trans-title-group></journal-title-group><issn pub-type="ppub">2226-1494</issn><issn pub-type="epub">2500-0373</issn><publisher><publisher-name>Университет ИТМО</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="doi">10.17586/2226-1494-2022-22-4-742-750</article-id><article-id custom-type="elpub" pub-id-type="custom">ntv-218</article-id><article-categories><subj-group subj-group-type="heading"><subject>Research Article</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="ru"><subject>КОМПЬЮТЕРНЫЕ СИСТЕМЫ И ИНФОРМАЦИОННЫЕ ТЕХНОЛОГИИ</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="en"><subject>COMPUTER SCIENCE</subject></subj-group></article-categories><title-group><article-title>Метод защиты нейронных сетей от компьютерных бэкдор-атак на основе идентификации триггеров закладок</article-title><trans-title-group xml:lang="en"><trans-title>A method for protecting neural networks from computer backdoor attacks based on the trigger identification</trans-title></trans-title-group></title-group><contrib-group><contrib contrib-type="author" corresp="yes"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0002-9955-2694</contrib-id><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Менисов</surname><given-names>А. Б.</given-names></name><name name-style="western" xml:lang="en"><surname>Menisov</surname><given-names>A. B.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Менисов Артем Бакытжанович — кандидат технических наук, докторант</p><p>sc 57220815185</p><p>Санкт-Петербург, 197198</p></bio><bio xml:lang="en"><p>Artem B. Menisov — PhD, Doctoral Student</p><p>sc 57220815185</p><p>Saint Petersburg, 197198</p></bio><email xlink:type="simple">vka@mil.ru</email><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0002-1764-1942</contrib-id><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Ломако</surname><given-names>А. Г.</given-names></name><name name-style="western" xml:lang="en"><surname>Lomako</surname><given-names>A. G.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Ломако Александр Григорьевич — доктор технических наук, профессор</p><p>sc 57188270500</p><p>Санкт-Петербург, 197198</p></bio><bio xml:lang="en"><p>Aleksandr G. Lomako — D. Sc., Full Professor</p><p>sc 57188270500</p><p>Saint Petersburg, 197198</p></bio><email xlink:type="simple">vka@mil.ru</email><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><contrib-id contrib-id-type="orcid">https://orcid.org/0000-0003-0283-9048</contrib-id><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Дудкин</surname><given-names>А. С.</given-names></name><name name-style="western" xml:lang="en"><surname>Dudkin</surname><given-names>A. S.</given-names></name></name-alternatives><bio xml:lang="ru"><p>Дудкин Андрей Сергеевич — кандидат технических наук, заместитель начальника кафедры</p><p>sc 57211979130</p><p>Санкт-Петербург, 197198</p></bio><bio xml:lang="en"><p>Andrey S. Dudkin — PhD, Deputy Head of Department</p><p>sc 57211979130</p><p>Saint Petersburg, 197198</p></bio><email xlink:type="simple">vka@mil.ru</email><xref ref-type="aff" rid="aff-1"/></contrib></contrib-group><aff-alternatives id="aff-1"><aff xml:lang="ru"><institution>Военно-космическая академия имени А.Ф.Можайского</institution><country>Россия</country></aff><aff xml:lang="en"><institution>Mozhaisky Military Aerospace Academy</institution><country>Russian Federation</country></aff></aff-alternatives><pub-date pub-type="collection"><year>2022</year></pub-date><pub-date pub-type="epub"><day>15</day><month>12</month><year>2024</year></pub-date><volume>22</volume><issue>4</issue><fpage>742</fpage><lpage>750</lpage><permissions><copyright-statement>Copyright &amp;#x00A9; Менисов А.Б., Ломако А.Г., Дудкин А.С., 2024</copyright-statement><copyright-year>2024</copyright-year><copyright-holder xml:lang="ru">Менисов А.Б., Ломако А.Г., Дудкин А.С.</copyright-holder><copyright-holder xml:lang="en">Menisov A.B., Lomako A.G., Dudkin A.S.</copyright-holder><license xml:lang="ru" license-type="creative-commons-attribution" xlink:href="https://creativecommons.org/licenses/by/4.0/" xlink:type="simple"><license-p>Данная работа распространяется под лицензией Creative Commons Attribution 4.0.</license-p></license><license xml:lang="en" license-type="creative-commons-attribution" xlink:href="https://creativecommons.org/licenses/by/4.0/" xlink:type="simple"><license-p>This work is licensed under a Creative Commons Attribution 4.0 License.</license-p></license></permissions><self-uri xlink:href="https://ntv.elpub.ru/jour/article/view/218">https://ntv.elpub.ru/jour/article/view/218</self-uri><abstract><p>Предмет исследования. Современные технологии разработки и эксплуатации нейронных сетей уязвимы для компьютерных атак с внедрением программных закладок (бэкдор). Программные закладки могут оставаться скрытыми неопределенное время, пока не будут активированы вводом модифицированных данных, содержащих триггеры. Такие закладки представляют непосредственную угрозу безопасности информации для всех компонентов системы искусственного интеллекта. Такие воздействия злоумышленников приводят к ухудшению качества или полному прекращению функционирования систем искусственного интеллекта. В работе предложен оригинальный метод защиты нейронных сетей, сущность которого состоит в создании базы ранжированных синтезированных триггеров закладок целевого класса бэкдор-атак. Метод. Предложенный метод защиты нейронных сетей реализован путем последовательности защитных действий: выявлении закладки, идентификации триггера и нейтрализации закладки. Основные результаты. На основе представленного метода разработано программно-алгоритмическое обеспечение испытаний нейронных сетей, позволяющее выявить и нейтрализовать закладки для осуществления компьютерных бэкдор-атак. Экспериментальные исследования проведены на различных архитектурах сверточных нейронных сетей, обученных на наборах данных, для таких объектов, как аэрофотоснимки (DOTA), рукописные цифры (MNIST) и фотографии лиц людей (LFW). Снижение эффективности бэкдор-атак (не более 3 %) и малые потери качества функционирования нейронных сетей (на 8–10 % от качества функционирования нейронной сети без закладки) показало успешность разработанного метода. Практическая значимость. Применение предложенного метода защиты нейронных сетей позволит специалистам по информационной безопасности целенаправленно противодействовать компьютерным бэкдор-атакам на системы искусственного интеллекта и создать новые автоматизированные средства защиты информации.</p></abstract><trans-abstract xml:lang="en"><p>Modern technologies for the development and operation of neural networks are vulnerable to computer attacks with the introduction of software backdoors. Program backdoors can remain hidden indefinitely until activated by input of modified data containing triggers. These backdoors pose a direct threat to the security of information for all components of the artificial intelligence system. Such influences of intruders lead to a deterioration in the quality or complete cessation of the functioning of artificial intelligence systems. This paper proposes an original method for protecting neural networks, the essence of which is to create a database of ranked synthesized backdoor’s triggers of the target class of backdoor attacks. The proposed method for protecting neural networks is implemented through a sequence of protective actions: detecting a backdoor, identifying a trigger, and neutralizing a backdoor. Based on the proposed method, software and algorithmic support for testing neural networks has been developed that allows you to identify and neutralize computer backdoor attacks. Experimental studies have been carried out on various dataset-trained convolutional neural network architectures for objects such as aerial photographs (DOTA), handwritten digits (MNIST), and photographs of human faces (LFW). The decrease in the effectiveness of backdoor attacks (no more than 3 %) and small losses in the quality of the functioning of neural networks (by 8–10 % of the quality of the functioning of a neural network without a backfill) showed the success of the developed method. The use of the developed method for protecting neural networks allows information security specialists to purposefully counteract computer backdoor attacks on artificial intelligence systems and develop automated information protection tools.</p></trans-abstract><kwd-group xml:lang="ru"><kwd>искусственный интеллект</kwd><kwd>искусственная нейронная сеть</kwd><kwd>информационная безопасность</kwd><kwd>компьютерные атаки</kwd><kwd>бэкдор</kwd><kwd>закладки в нейронных сетях</kwd><kwd>синтезированные триггеры</kwd></kwd-group><kwd-group xml:lang="en"><kwd>artificial intelligence</kwd><kwd>artificial neural network</kwd><kwd>information security</kwd><kwd>computer attacks</kwd><kwd>backdoor</kwd><kwd>backdoors in neural networks</kwd><kwd>synthesized triggers</kwd></kwd-group><funding-group><funding-statement xml:lang="ru">Работа выполнена в рамках гранта Президента Российской Федерации для государственной поддержки молодых российских ученых — кандидатов наук МК-2485.2022.4.</funding-statement><funding-statement xml:lang="en">The work was carried out within the framework of the grant of the President of the Russian Federation for state support of young Russian scientists — candidates of sciences MK-2485.2022.4.</funding-statement></funding-group></article-meta></front><back><ref-list><title>References</title><ref id="cit1"><label>1</label><citation-alternatives><mixed-citation xml:lang="ru">Буханов Д.Г., Поляков В.М., Редькина М.А. Обнаружение вредоносного программного обеспечения с использованием искусственной нейронной сети на основе адаптивно-резонансной теории // Прикладная дискретная математика. 2021. № 52. С. 69–82. https://doi.org/10.17223/20710410/52/4</mixed-citation><mixed-citation xml:lang="en">Bukhanov D.G., Polyakov V.M., Redkina M.A. Detection of Malware using an artificial neural network based on adaptive resonant theory. Prikladnaya Diskretnaya Matematika, 2021, no. 52, pp. 69–82. (in Russian). https://doi.org/10.17223/20710410/52/4</mixed-citation></citation-alternatives></ref><ref id="cit2"><label>2</label><citation-alternatives><mixed-citation xml:lang="ru">Massarelli L., Di Luna G.A., Petroni F., Querzoni L., Baldoni R. Investigating graph embedding neural networks with unsupervised features extraction for binary analysis // Proc. of the 2nd Workshop on Binary Analysis Research (BAR). 2019. https://dx.doi.org/10.14722/bar.2019.23020</mixed-citation><mixed-citation xml:lang="en">Massarelli L., Di Luna G.A., Petroni F., Querzoni L., Baldoni R. Investigating graph embedding neural networks with unsupervised features extraction for binary analysis. Proc. of the 2nd Workshop on Binary Analysis Research (BAR), 2019, https://dx.doi.org/10.14722/bar.2019.23020</mixed-citation></citation-alternatives></ref><ref id="cit3"><label>3</label><citation-alternatives><mixed-citation xml:lang="ru">Забелина В.А., Савченко Г.А., Черненький И.М., Силантьева Е.Ю. Обнаружение Интернет-атак с помощью нейронной сети // Динамика сложных систем-XXI век. 2021. Т. 15. № 2. С. 39–47. https://doi.org/10.18127/j19997493-202102-04</mixed-citation><mixed-citation xml:lang="en">Zabelina V.A., Savchenko G.A., Chernenky I.M., Silantieva E.Yu. Detecting internet attacks using a neural network. Dynamics of Complex Systems — XXI century, 2021, vol. 15, no. 2, pp. 39–47. (in Russian). https://doi.org/10.18127/j19997493-202102-04</mixed-citation></citation-alternatives></ref><ref id="cit4"><label>4</label><citation-alternatives><mixed-citation xml:lang="ru">Архипова А.Б., Поляков П.А. Методология построения нейронной нечеткой сети в области информационной безопасности // Безопасность цифровых технологий. 2021. № 3. С. 43–56. https://doi.org/10.17212/2782-2230-2021-3-43-56</mixed-citation><mixed-citation xml:lang="en">Arkhipova A.B., Polyakov P.A. Methodology for constructing a neural fuzzy network in the field of information security. Digital Technology Security, 2021, no. 3, pp. 43–56. (in Russian). https://doi.org/10.17212/2782-2230-2021-3-43-56</mixed-citation></citation-alternatives></ref><ref id="cit5"><label>5</label><citation-alternatives><mixed-citation xml:lang="ru">Спицын В.Г., Цой Ю.Р. Эволюционирующие искусственные нейронные сети // Сборник трудов IV Всероссийской конференции студентов, аспирантов и молодых ученых «Молодежь и современные информационные технологии», Томск, 28 февраля — 2 марта, 2006 г. Томск, 2006. C. 411–413.</mixed-citation><mixed-citation xml:lang="en">Spitcyn V.G., Tcoi Iu.R. Evolving artificial neural networks. Proc. of the IV All-Russian conference of students, graduate students and young scientists “Youth and Modern Information Technologies”, Tomsk, February 28 — March 2, 2006, Tomsk, 2006, pp. 411–413. (in Russian)</mixed-citation></citation-alternatives></ref><ref id="cit6"><label>6</label><citation-alternatives><mixed-citation xml:lang="ru">Мак-Каллок У.С., Питтс В. Логическое исчисление идей, относящихся к нервной активности // Автоматы / под ред. К.Э. Шеннона и Дж. Маккарти. М.: Иностранная литература, 1956. С. 363–384.</mixed-citation><mixed-citation xml:lang="en">McCulloch W.S., Pitts V. A logical calculus of the ideas immanent in nervous activity. Automata studies. Ed. by. C.E. Shannon and McCarthy. Princeton - New Jersey, Princeton univ. press, 1956.</mixed-citation></citation-alternatives></ref><ref id="cit7"><label>7</label><citation-alternatives><mixed-citation xml:lang="ru">Шевская Н.В. Объяснимый искусственный интеллект и методы интерпретации результатов // Моделирование, оптимизация и информационные технологии. 2021. Т. 9. № 2. С. 22–23. https:// doi.org/10.26102/2310-6018/2021.33.2.024</mixed-citation><mixed-citation xml:lang="en">Shevskaya N.V. Explainable artificial intelligence and methods for interpreting results. Modeling, Optimization and Information Technology, 2021, vol. 9, no. 2, pp. 22–23. (in Russian). https://doi.org/10.26102/2310-6018/2021.33.2.024</mixed-citation></citation-alternatives></ref><ref id="cit8"><label>8</label><citation-alternatives><mixed-citation xml:lang="ru">Xu Q., Arafin M.T., Qu G. Security of neural networks from hardware perspective: A survey and beyond // Proc. of the 26th Asia and South Pacific Design Automation Conference (ASP-DAC). 2021. P. 449–454. https://doi.org/10.1145/3394885.3431639</mixed-citation><mixed-citation xml:lang="en">Xu Q., Arafin M.T., Qu G. Security of neural networks from hardware perspective: A survey and beyond. Proc. of the 26th Asia and South Pacific Design Automation Conference (ASP-DAC), 2021, pp. 449–454. https://doi.org/10.1145/3394885.3431639</mixed-citation></citation-alternatives></ref><ref id="cit9"><label>9</label><citation-alternatives><mixed-citation xml:lang="ru">Kravets V., Javidi B., Stern A. Defending deep neural networks from adversarial attacks on three-dimensional images by compressive sensing // Proc. of the 3D Image Acquisition and Display: Technology, Perception and Applications. 2021.</mixed-citation><mixed-citation xml:lang="en">Kravets V., Javidi B., Stern A. Defending deep neural networks from adversarial attacks on three-dimensional images by compressive sensing. Proc. of the 3D Image Acquisition and Display: Technology, Perception and Applications, 2021.</mixed-citation></citation-alternatives></ref><ref id="cit10"><label>10</label><citation-alternatives><mixed-citation xml:lang="ru">Liu Y., Ma S., Aafer Y., Lee W.-C., Zhai J. Trojaning attack on neural networks: Report 17-002. 2017.</mixed-citation><mixed-citation xml:lang="en">Liu Y., Ma S., Aafer Y., Lee W.-C., Zhai J. Trojaning attack on neural networks. Report 17-002, 2017.</mixed-citation></citation-alternatives></ref><ref id="cit11"><label>11</label><citation-alternatives><mixed-citation xml:lang="ru">Chen X., Liu C., Li B., Lu K., Song D. Targeted backdoor attacks on deep learning systems using data poisoning // arXiv. 2017. arXiv:1712.05526. https://doi.org/10.48550/arXiv.1712.05526</mixed-citation><mixed-citation xml:lang="en">Chen X., Liu C., Li B., Lu K., Song D. Targeted backdoor attacks on deep learning systems using data poisoning. arXiv, 2017, arXiv:1712.05526. https://doi.org/10.48550/arXiv.1712.05526</mixed-citation></citation-alternatives></ref><ref id="cit12"><label>12</label><citation-alternatives><mixed-citation xml:lang="ru">Li W., Yu J., Ning X., Wang P., Wei Q., Wang Y., Yang H. Hu-Fu: Hardware and software collaborative attack framework against neural networks // Proc. of the 17th IEEE Computer Society Annual Symposium on VLSI (ISVLSI). 2018. P. 482–487. https://doi.org/10.1109/ISVLSI.2018.00093</mixed-citation><mixed-citation xml:lang="en">Li W., Yu J., Ning X., Wang P., Wei Q., Wang Y., Yang H. Hu-Fu: Hardware and software collaborative attack framework against neural networks. Proc. of the 17th IEEE Computer Society Annual Symposium on VLSI (ISVLSI), 2018, pp. 482–487. https://doi.org/10.1109/ISVLSI.2018.00093</mixed-citation></citation-alternatives></ref><ref id="cit13"><label>13</label><citation-alternatives><mixed-citation xml:lang="ru">Gong X., Chen Y., Wang Q., Huang H., Meng L., Shen C., Zhang Q. Defense-resistant backdoor attacks against deep neural networks in outsourced cloud environment // IEEE Journal on Selected Areas in Communications. 2021. V. 39. N 8. P. 2617–2631. https://doi.org/10.1109/JSAC.2021.3087237</mixed-citation><mixed-citation xml:lang="en">Gong X., Chen Y., Wang Q., Huang H., Meng L., Shen C., Zhang Q. Defense-resistant backdoor attacks against deep neural networks in outsourced cloud environment. IEEE Journal on Selected Areas in Communications, 2021, vol. 39, no. 8, pp. 2617–2631. https://doi.org/10.1109/JSAC.2021.3087237</mixed-citation></citation-alternatives></ref><ref id="cit14"><label>14</label><citation-alternatives><mixed-citation xml:lang="ru">Wenger E., Passananti J., Bhagoji A.N., Yao Y., Zheng H., Zhao B.Y. Backdoor attacks against deep learning systems in the physical world // Proc. of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2021. P. 6202–6211. https://doi.org/10.1109/CVPR46437.2021.00614</mixed-citation><mixed-citation xml:lang="en">Wenger E., Passananti J., Bhagoji A.N., Yao Y., Zheng H., Zhao B.Y. Backdoor attacks against deep learning systems in the physical world. Proc. of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021, pp. 6202–6211. https://doi.org/10.1109/CVPR46437.2021.00614</mixed-citation></citation-alternatives></ref><ref id="cit15"><label>15</label><citation-alternatives><mixed-citation xml:lang="ru">Shahroudnejad A. A survey on understanding, visualizations, and explanation of deep neural networks // arXiv. 2021. arXiv:2102.01792. https://doi.org/10.48550/arXiv.2102.01792</mixed-citation><mixed-citation xml:lang="en">Shahroudnejad A. A survey on understanding, visualizations, and explanation of deep neural networks. arXiv, 2021, arXiv:2102.01792. https://doi.org/10.48550/arXiv.2102.01792</mixed-citation></citation-alternatives></ref><ref id="cit16"><label>16</label><citation-alternatives><mixed-citation xml:lang="ru">Wang B., Yao Y., Shan Sh., Li H., Viswanath B., Zheng H., Zhao B.Y. Neural cleanse: Identifying and mitigating backdoor attacks in neural networks // Proc. of the 40th IEEE Symposium on Security and Privacy (SP). 2019. P. 707–723. https://doi.org/10.1109/SP.2019.00031</mixed-citation><mixed-citation xml:lang="en">Wang B., Yao Y., Shan Sh., Li H., Viswanath B., Zheng H., Zhao B.Y. Neural cleanse: Identifying and mitigating backdoor attacks in neural networks. Proc. of the 40th IEEE Symposium on Security and Privacy (SP), 2019, pp. 707–723. https://doi.org/10.1109/SP.2019.00031</mixed-citation></citation-alternatives></ref><ref id="cit17"><label>17</label><citation-alternatives><mixed-citation xml:lang="ru">Xia G.-S., Bai X., Ding J., Zhu Z., Belongie S., Luo J., Datcu M., Pelillo M., Zhang L. DOTA: A large-scale dataset for object detection in aerial images // Proc. of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2018. P. 3974–3983. https://doi.org/10.1109/CVPR.2018.00418</mixed-citation><mixed-citation xml:lang="en">Xia G.-S., Bai X., Ding J., Zhu Z., Belongie S., Luo J., Datcu M., Pelillo M., Zhang L. DOTA: A large-scale dataset for object detection in aerial images. Proc. of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018, pp. 3974–3983. https://doi.org/10.1109/CVPR.2018.00418</mixed-citation></citation-alternatives></ref><ref id="cit18"><label>18</label><citation-alternatives><mixed-citation xml:lang="ru">Deng L. The MNIST database of handwritten digit images for machine learning research // IEEE Signal Processing Magazine. 2012. V. 29. N 6. P. 141–142. https://doi.org/10.1109/MSP.2012.2211477</mixed-citation><mixed-citation xml:lang="en">Deng L. The MNIST database of handwritten digit images for machine learning research. IEEE Signal Processing Magazine, 2012, vol. 29, no. 6, pp. 141–142. https://doi.org/10.1109/MSP.2012.2211477</mixed-citation></citation-alternatives></ref><ref id="cit19"><label>19</label><citation-alternatives><mixed-citation xml:lang="ru">Huang G.B., Mattar M., Berg T., Learned-Miller E. Labeled faces in the wild: A database forstudying face recognition in unconstrained environments // Proc. of the Workshop on Faces in ‘Real-Life’ Images: Detection, Alignment, and Recognition. 2008.</mixed-citation><mixed-citation xml:lang="en">Huang G.B., Mattar M., Berg T., Learned-Miller E. Labeled faces in the wild: A database forstudying face recognition in unconstrained environments. Proc. of the Workshop on Faces in ‘Real-Life’ Images: Detection, Alignment, and Recognition, 2008.</mixed-citation></citation-alternatives></ref><ref id="cit20"><label>20</label><citation-alternatives><mixed-citation xml:lang="ru">Wang J., Xiao H., Chen L., Xing J., Pan Z., Luo R., Cai X. Integrating weighted feature fusion and the spatial attention module with convolutional neural networks for automatic aircraft detection from SAR images // Remote Sensing. 2021. V. 13. N 5. P. 910. https://doi.org/10.3390/rs13050910</mixed-citation><mixed-citation xml:lang="en">Wang J., Xiao H., Chen L., Xing J., Pan Z., Luo R., Cai X. Integrating weighted feature fusion and the spatial attention module with convolutional neural networks for automatic aircraft detection from SAR images. Remote Sensing, 2021, vol. 13, no. 5, pp. 910. https://doi.org/10.3390/rs13050910</mixed-citation></citation-alternatives></ref><ref id="cit21"><label>21</label><citation-alternatives><mixed-citation xml:lang="ru">An S., Lee M., Park S., Yang H., Soet J. An ensemble of simple convolutional neural network models for MNIST digit recognition // arXiv. 2020. arXiv:2008.10400. https://doi.org/10.48550/arXiv.2008.10400</mixed-citation><mixed-citation xml:lang="en">An S., Lee M., Park S., Yang H., Soet J. An ensemble of simple convolutional neural network models for MNIST digit recognition. arXiv, 2020, arXiv:2008.10400. https://doi.org/10.48550/arXiv.2008.10400</mixed-citation></citation-alternatives></ref><ref id="cit22"><label>22</label><citation-alternatives><mixed-citation xml:lang="ru">Yan M., Zhao M., Xu Z., Zhang Q., Wang G., Su Z. VarGFaceNet: An efficient variable group convolutional neural network for lightweight face recognition // Proc. of the 17th IEEE/CVF International Conference on Computer Vision Workshops (ICCVW). 2019. P. 2647–2654. https://doi.org/10.1109/ICCVW.2019.00323</mixed-citation><mixed-citation xml:lang="en">Yan M., Zhao M., Xu Z., Zhang Q., Wang G., Su Z. VarGFaceNet: An efficient variable group convolutional neural network for lightweight face recognition. Proc. of the 17th IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), 2019, pp. 2647–2654. https://doi.org/10.1109/ICCVW.2019.00323</mixed-citation></citation-alternatives></ref><ref id="cit23"><label>23</label><citation-alternatives><mixed-citation xml:lang="ru">Liu X., Li F., Wen B., Li Q. Removing backdoor-based watermarks in neural networks with limited data // Proc. of the 25th International Conference on Pattern Recognition (ICPR). 2021. P. 10149–10156. https://doi.org/10.1109/ICPR48806.2021.9412684</mixed-citation><mixed-citation xml:lang="en">Liu X., Li F., Wen B., Li Q. Removing backdoor-based watermarks in neural networks with limited data. Proc. of the 25th International Conference on Pattern Recognition (ICPR), 2021, pp. 10149–10156. https://doi.org/10.1109/ICPR48806.2021.9412684</mixed-citation></citation-alternatives></ref><ref id="cit24"><label>24</label><citation-alternatives><mixed-citation xml:lang="ru">Kaviani S., Sohn I. Defense against neural trojan attacks: A survey // Neurocomputing. 2021. V. 423. P. 651–667. https://doi.org/10.1016/j.neucom.2020.07.133</mixed-citation><mixed-citation xml:lang="en">Kaviani S., Sohn I. Defense against neural trojan attacks: A survey. Neurocomputing, 2021, vol. 423, pp. 651–667. https://doi.org/10.1016/j.neucom.2020.07.133</mixed-citation></citation-alternatives></ref></ref-list><fn-group><fn fn-type="conflict"><p>The authors declare that there are no conflicts of interest present.</p></fn></fn-group></back></article>
