The 1st ACNS Workshop on Security and Privacy of Federated Learning


S&P-FL

June 19-22, 2023

Kyoto, Japan

in conjunction with ACNS 2023 (June 19-22, 2023)

Call for Papers

The success of deep learning stems from availability of big training data and massive computation power. However, in many applications, training data are generated by individuals or organizations, who hesitate to share their data that expose privacy. Federated learning has been proposed to enable distributed computing nodes to collaboratively train models without exposing their own data. Its basic idea is to let these computing nodes train local models using their own data, respectively, and then upload the local models, instead of raw data, to a logically centralized parameter server that synthesizes a global model.

Despite the promise in data protection, federated learning faces new security and privacy threats. Some recent research has shown that it is possible to infer training data information by observing shared models. In addition, there is strong desire to protect models because model design needs significant investment and they are treated as important digital assets. However, models are exposed to everyone in the default design of federated learning. Furthermore, malicious participants may exist in federated learning and they would compromise the whole learning process by sharing wrong models. There also may exist free-riders that enjoy the shared model, without making contributions.

Addressing the above challenges of federated learning security and privacy needs significant research efforts on theories, algorithms, architecture, and experiences of system deployment and maintenance. Therefore, this workshop aims to offer a platform for researchers from both academia and industry to publish recent research findings and to discuss opportunities, challenges and solutions related to security and privacy of federated learning.

Topics of interests include, but are not limited to:

  • Security and privacy analysis of federated learning

  • Data privacy enhancement of federated learning

  • Model protection of federated learning

  • Secure multi-party computation for federated learning

  • Homomorphic encryption of federated learning

  • Differential privacy of federated learning

  • Tradeoff between privacy and efficiency of federated learning

  • Software system security of federated learning

  • Hardware security of federated learning

  • Network security of federated learning

  • Quantum security for federated learning

  • Blockchain for security and privacy protection of federated learning

  • Emerging threat and attack for federated learning

Important Dates

Submission Instructions

Submissions must be anonymous, with no author names, affiliations, acknowledgments, or obvious references.

All submissions must follow the original LNCS format with a page limit of 18 pages (including references) . Submissions not meeting the submission guidelines risk rejection without consideration of their merits.

The authors are invited to submit the papers using EasyChair submission system . The proceedings of the workshop will be published by Springer in the LNCS series. Best paper award will be selected paper based on the reviews by the committee.

Program Chairs

PC Members