Training Description
Key Takeaways
- Focuses on secure and efficient federated learning (FL) for distributed AI
- Explores privacy-preserving techniques and robust aggregation methods
- Addresses challenges in heterogeneous and resource-constrained environments
- Bridges theoretical research and real-world FL deployments
- Targets academic researchers, industry practitioners, and technical leaders in AI/ML
The Workshop on Secure and Efficient Federated Learning (FL-AsiaCCS 2026) is a specialized event dedicated to advancing federated learning, a collaborative machine learning approach that preserves data privacy across decentralized devices. This workshop brings together leading researchers and industry professionals to discuss the latest developments, challenges, and practical applications in the field.
Overview of the Workshop
The event centers on federated learning, emphasizing the need for secure, efficient, and scalable systems as data diversity and volume continue to grow. With the increasing adoption of large-scale models, such as large language models, the workshop highlights the importance of privacy and security in distributed AI environments. Attendees can expect a comprehensive program featuring keynotes, technical sessions, and tutorials that cover both foundational research and practical deployments.
Participants will gain insights into the latest advances in federated learning, including communication efficiency, coded FL, and privacy-preserving techniques. The workshop also explores robust aggregation methods, trusted execution environments, and defenses against security attacks, ensuring that FL systems remain resilient and trustworthy.
Main Topics and Technologies
Key topics include scalable and robust federated learning, FL in heterogeneous networks, and FL for large language models. The event delves into quantum secure aggregation, verifiable FL, and practical frameworks such as Ed-Fed and Flower. These discussions are designed to address real-world challenges, from defending against adversarial attacks to managing system faults and device heterogeneity.
Notable organizations involved include Nanyang Technological University, KTH Royal Institute of Technology, Technical University of Denmark, Technical University of Munich, and TrustFUL. The workshop also features hands-on tutorials and online plenary talks, providing a well-rounded experience for all attendees.
Audience and Value
This workshop is tailored for academic researchers, industry practitioners, and technical leaders in artificial intelligence, machine learning, cybersecurity, and distributed systems. Attendees include professors, research scientists, engineers, PhD students, and R&D managers from universities, technology companies, and research institutes.
The event addresses critical challenges such as ensuring data privacy, improving training efficiency, and defending against adversarial threats in distributed AI systems. By fostering collaboration and knowledge exchange, the workshop aims to bridge the gap between theoretical frameworks and practical edge deployments, supporting the advancement of secure and efficient federated learning.