BEiNG WISE

Polytechnic University of Tirana, Albania

Venue: Universiteti Politeknik i Tiranes, Bulevardi Dëshmorët e Kombit Nr. 4, Tiranë 1000, Albania

The COST Action CA22104  BEiNG-WISE, is pleased in launching the CALL for trainees for the Training School titled: The Double Nature of Machine Learning (ML)/ Artificial Intelligence (AI) in Cybersecurity.

This Training School has the main objective to present some advanced aspects related to cybersecurity and Machine Learning (ML)/Artificial Intelligence (AI) approaches. Specific thematic related to explainability, reliability will be considered in relation to both cyber-security and ML/AI methods. These aspects will be also related to ethics and legal factors, and a cybercrime point of view will be also addressed.

The main topics include:

• Explainable AI for Intrusion Detection in Next-Generation Wireless Networks
• Behavioral-Based Anomaly Detection Using XAI for Cybersecurity in 6G Networks
• Enhancing Reliability of Federated Learning Models for Wireless Network Security
• Explainable and Trustworthy AI for User Behavior Analysis in Cybersecurity

The primary goal of the school is to provide participants with an in-depth understanding of cutting-edge topics at the intersection of cybersecurity and Explainable AI (XAI) within the context of next-generation wireless networks, including 6G. As wireless communication technologies evolve, the integration of AI-driven solutions is becoming indispensable for securing these networks. However, the interpretability and trustworthiness of AI models remain a major challenge. This school aims to equip participants with the knowledge and skills necessary to address these challenges through focused discussions, practical applications, and collaborative research.

A key focus of the school is to investigate methods for enhancing the interpretability and reliability of AI-driven intrusion detection systems (IDS) tailored for next-generation wireless networks. As AI-based security solutions gain traction, it is crucial that their decision-making processes are transparent and understandable to human operators. Participants will explore state-of-the-art techniques in explainable AI for IDS, examine methods for balancing detection accuracy and interpretability, develop strategies to enhance human trust and confidence in AI-driven security solutions, and identify challenges and limitations of explainability in real-world deployment scenarios.

Another important aspect of the program is developing behavioral anomaly detection models that leverage XAI to detect deviations from normal behavior patterns in 6G networks in real-time. Behavioral-based models offer a promising approach to identifying unknown and evolving threats. Participants will gain an understanding of the unique security challenges in 6G networks, develop behavioral profiling techniques using XAI, enhance real-time detection capabilities to improve response times, and evaluate the performance of XAI-based anomaly detection in dynamic network environments.

Improving the reliability and robustness of federated learning models utilized for wireless network security is another major goal of the school. Federated learning offers a decentralized approach to AI model training, but it introduces challenges related to explainability, security, and data integrity. The school will investigate techniques to enhance model accuracy and robustness in federated environments, address challenges such as adversarial attacks and data poisoning, improve the interpretability of federated learning models to foster trust, and develop frameworks to ensure data privacy while maintaining high detection rates.

The final area of focus involves designing explainable AI models to analyze user behavior in wireless networks, aiming to detect insider threats and malicious activities reliably. User behavior analysis can provide critical insights into potential security risks, but it must be conducted in a transparent and ethical manner. Participants will work on developing models that provide interpretable insights into user activities, identifying patterns indicative of insider threats and advanced persistent threats, establishing frameworks for ethical user behavior analysis, and ensuring compliance with data privacy regulations while maintaining security.

By the end of this school, participants will have gained a comprehensive understanding of how explainable AI can enhance cybersecurity in next-generation wireless networks. They will be equipped with the necessary skills to develop interpretable AI models, improve reliability in federated learning applications, and design human-centric security solutions. This knowledge will empower them to contribute to the advancement of secure, trustworthy, and explainable AI-driven cybersecurity frameworks in the evolving wireless communication landscape.

There are several main specific objectives of BEiNG-WISE COST Action and the Training School:

–        Promote Interdisciplinary Collaboration: Encourage interdisciplinary collaboration among participants from diverse backgrounds, including doctoral students, researchers, policymakers, and industry professionals. By fostering collaboration, the training program aims to facilitate knowledge exchange, foster innovative thinking, and inspire cross-sector partnerships.

–        Develop Analytical and Critical Thinking Skills: Enhance participants’ analytical and critical thinking skills by engaging them in discussions, case studies, and interactive sessions. The program aims to empower participants to evaluate the social, economic, ethical, and policy dimensions of Cybersecurity and develop informed perspectives.

–        Equip Participants with Practical Knowledge and Tools: Provide participants with practical knowledge and tools that enable them to navigate the complex  landscape of Cyber Security in wireless networks. This includes understanding policy frameworks, leveraging emerging technologies, and developing skills and capabilities relevant to the digital era.

–        Facilitate Networking and Collaboration Opportunities: Create a conducive environment for networking and collaboration among participants, guest speakers, and experts in the field. The program intends to foster connections leading to future research collaborations, joint projects, and professional development opportunities in the digitalization domain.

–        By achieving these goals, the training school aspires to empower participants to become well-rounded scholars and practitioners who can contribute to advancing policies, technologies, and challenges for Cyber Security in wireless networks.

The BEiNG-WISE Training School is a multidisciplinary event open to professionals, researchers, and Ph.D. students with a professional interest in the topic. Any interested applicant must complete the Application Form.

Criteria for selected trainees

Applicants shall be engaged in an official research program as a PhD Student or postdoctoral fellow or can be employed by, or affiliated to, an institution, organization, or legal entity which has within its remit a clear association with the topics of the COST Action. Also undergraduate and graduate students will be taken into consideration.

Trainees eligible for reimbursement are from COST Full Member Countries, COST Near Neighbor Countries (NNC) or Approved European RTD Organizations.

The selected international applicants (trainees) will benefit from a Trainee Grant to participate in the activities of the Training School from the 4^th to the 6^th of June 2025. This grant is expected to cover the long-distance travel expenses (by train, ferry, bus, plane, or car). Furthermore, the international trainees will receive a daily allowance, which is expected to cover the accommodation in Tirana, meals and local transportation. Each application will be individually evaluated along with the total grant awarded for the 3 days.

The selected applicants (trainees) from Albania will benefit from a Trainee Grant to participate in the activities of the Training School based on the distance to and from Tirana and the accommodation needs. This grant will be individually discussed and agreed.

Selected trainees shall create an e-COST profile at https://e-services.cost.eu  including bank details prior to accepting their e-COST invitation. They shall submit via e-COST a completed online travel reimbursement request within 30 calendar days after the end date of the approved activity.

We are excited that you are part of this program and at the opportunities it will provide to you!

The training will revolve around the following topics:

• Explainable AI for Intrusion Detection in Next-Generation Wireless Networks
• Behavioral-Based Anomaly Detection Using XAI for Cybersecurity in 6G Networks
• Enhancing Reliability of Federated Learning Models for Wireless Network Security
• Explainable and Trustworthy AI for User Behavior Analysis in Cybersecurity

In an era defined by rapid technological advancement and digital connectivity, the training school is dedicated to exploring the intricate interplay between cybersecurity and AI/ML This school is designed to equip participants with comprehensive insights into the synergy between cybersecurity and Explainable AI (XAI) within next-generation wireless networks. As AI-driven security measures become increasingly integrated into wireless systems, it is essential to ensure their transparency and dependability. The program will delve into strategies for improving the interpretability of AI-based intrusion detection systems, developing behavioral anomaly detection frameworks for 6G networks, and strengthening the reliability of federated learning models in cybersecurity applications. Attendees will examine methods to optimize detection performance while maintaining clarity, uncover security threats through behavioral analysis, and address critical challenges such as adversarial threats and data security. Furthermore, the program will cover ethical and regulatory aspects related to analyzing user behavior to identify potential threats while safeguarding privacy and compliance. By the conclusion of the school, participants will possess the expertise needed to develop transparent AI models and contribute to the creation of more secure and explainable AI-driven cybersecurity solutions in evolving wireless networks.

Central to the training school’s success are the trainers  – an exceptional group of scholars, researchers, and industry professionals who bring unparalleled expertise and experience to the classroom. Committed to excellence in teaching and research, the trainers will serve as mentors, guides, and collaborators, nurturing the intellectual growth and professional development of the trainees.

The school will foster a culture of innovation and inquiry, where trainees have the opportunity to explore cutting-edge research topics, collaborate with peers and faculty, and make meaningful contributions to the advancement of cybersecurity knowledge. From empirical studies on human factors in wireless security to theoretical investigations of optimal security approaches, the school will discuss a range of topics.

The training school is more than just an educational session – it should be a vibrant community of scholars, practitioners, and thought leaders united by a shared commitment to excellence in cybersecurity. Through collaboration, mentorship, and networking opportunities, the school will foster a sense of camaraderie and mutual support that enriches the educational experience and extends beyond the classroom.

•       Applications opened: 1st February 2025
•       Applications submission deadline: 1^st of March 2025
•       Confirmation to the selected trainees: 15^th of March 2025
•       Deadline for e-Cost registration and confirmation: 31 March 2025
•       Training School days: 4-6 June 2025