Cybersecurity and Artificial Intelligence (CS&AI) Research Group

About Us

The CS&AI research group focuses on the interconnection of cybersecurity and artificial intelligence. On the one hand, it deals with the applications of artificial intelligence and machine learning methods in the field of cybersecurity, and on the other hand, it is dedicated to the security of artificial intelligence models. The group's goal is to contribute to the development of innovative solutions for the detection and prevention of cyber threats, protection against adversarial attacks, analyze the latest malware attack techniques, and disseminate knowledge in this area through publications and cooperation with the academic and industrial sectors.

People

• 🔹Mgr. Martin Jureček, Ph.D. - Head (Google Scholar)
• 🔹prof. Ing. Róbert Lórencz, CSc. - Deputy head (Google Scholar)
• 🔹Ing. Jan Dolejš - Doctoral student (UserMap of CTU)
• 🔹Mgr. Olha Jurečková - Doctoral student (Google Scholar)
• 🔹Ing. Matouš Kozák - Doctoral student (Google Scholar)
• 🔹Ing. Eliška Krátká - Doctoral student (UserMap of CTU)
• 🔹Ing. Benjamín Peraus - Doctoral student (UserMap of CTU)

Projects

• 🔹 Adversarial Learning Techniques for Malware Detection [3, 4, 7, 10]
  Exploring adversarial machine learning techniques to generate and detect evasive malware samples. The goal is to improve robustness by developing classifiers resistant to adversarial attacks.
• 🔹 Defense Against Adversarial Attacks for Malware Detection To be published in May
  Investigating defense mechanisms against adversarial malware designed to bypass detection systems. The focus is on adversarial training, input transformations, and ensemble learning methods.
• 🔹 Classification and Clustering of Malware Families Based on Machine Learning [5, 6, 11]
  Developing machine learning models to classify and cluster malware families efficiently. This research aims to improve zero-day malware detection and enhance threat intelligence.
🔹 Machine Learning for Malware Detection and Cybersecurity Challenges [2, 8, 9, 12, 13, 14, 15]
Designing machine learning models for malware detection, utilizing mainly static analysis to identify malicious patterns in executable files. The goal is to enhance automated threat detection and improve cybersecurity defenses by leveraging deep learning techniques.
• 🔹 Quantum Computing Methods for Malware Classification To be published in May
  Exploring the potential of quantum computing to improve malware classification. The project focuses on leveraging quantum algorithms for pattern recognition and anomaly detection in cybersecurity.
• 🔹 Algebraic Cryptanalysis Accelerated by Data Mining Methods [1]
  Utilizing data mining techniques to enhance algebraic cryptanalysis of symmetric ciphers. The research focuses on reducing polynomial systems to accelerate cryptanalytic attacks.

Publications

• 🔹[1] Reducing overdefined systems of polynomial equations derived from small scale variants of the AES via data mining methods - Berušková, J., Jureček, M., & Jurečková, O. (2024). Journal of Computer Virology and Hacking Techniques, 20(4), 885-900.
• 🔹[2] Social media bot detection using Dropout-GAN - Shukla, A., Jureček, M., Stamp, M. (2024). Journal of Computer Virology and Hacking Techniques, 20(4), 669-680.
• 🔹[3] A comparison of adversarial malware generators - Louthánová, P., Kozák, M., Jureček, M., Stamp, M., Di Troia, F. (2024). Journal of Computer Virology and Hacking Techniques, 20(4), 623-639.
• 🔹[4] Creating valid adversarial examples of malware - Kozák, M., Jureček, M., Stamp, M., Di Troia, F. (2024). Journal of Computer Virology and Hacking Techniques, 20(4), 607-621.
• 🔹[5] Classification and online clustering of zero-day malware - Jurečková, O., Jureček, M., Stamp, M., Di Troia, F., Lórencz, R. (2024). Journal of Computer Virology and Hacking Techniques, 20(4), 579-592.
• 🔹[6] A natural language processing approach to Malware classification - Mehta, R., Jurečková, O., Stamp, M. (2024). Journal of Computer Virology and Hacking Techniques, 20(1), 173-184.
• 🔹[7] Combining Generators of Adversarial Malware Examples to Increase Evasion Rate - Kozák, M., Jureček, M. (2023). 20th International Conference on Security and Cryptography (SECRYPT), 778-786 .
• 🔹[8] Parallel Instance Filtering for Malware Detection - Jureček, M., Jurečková, O. (2022). 48th Euromicro Conference on Software Engineering and Advanced Applications (SEAA), 13-20 .
• 🔹[9] Interpretability of Machine Learning-Based Results of Malware Detection Using a Set of Rules - Dolejš, J., Jureček, M. (2022). In Artificial Intelligence for Cybersecurity, 107-136.
• 🔹[10] Generation of adversarial malware and benign examples using reinforcement learning - Kozák, M., Jureček, M., Lórencz, R. (2022). In Artificial Intelligence for Cybersecurity, 3-25.
• 🔹[11] Application of Distance Metric Learning to Automated Malware Detection - Jureček, M., Lórencz, R. (2021). IEEE Access, 9, 96151-96165.
• 🔹[12] Improving Classification of Malware Families using Learning a Distance Metric - Jureček, M., Jurečková, O., Lórencz, R. (2021). 7th International Conference on Information Systems Security and Privacy (ICISSP), 643-652.
• 🔹[13] Representation of PE Files using LSTM Networks - Jureček, M., Kozák, M. (2021). 7th International Conference on Information Systems Security and Privacy (ICISSP), 516-525.
• 🔹[14] Distance Metric Learning using Particle Swarm Optimization to Improve Static Malware Detection - Jureček, M., Lórencz, R. (2020). 6th International Conference on Information Systems Security and Privacy (ICISSP), 516-525.
• 🔹[15] Malware detection using a heterogeneous distance function - Jureček, M., Lórencz, R. (2018). Computing and Informatics, 37(3), 759-780.

Contact Us

🔹Address:	Department of Information Security, Faculty of Information Technology, Czech Technical University in Prague, Thákurova 9, 160 00 Prague 6, Czech Republic
🔹Office:	TH:A-1137
🔹Email:	martin.jurecek (here at) fit.cvut.cz
🔹Phone:	+420-22435-9858

Location