Part of INNS International Joint Conference on Neural Networks (IJCNN) 2025
About the Special Session
Privacy-Preserving Machine and Deep Learning (PP-MDL) is an emerging research area focused on enabling inference and training of Machine and Deep Learning (ML and DL) models in ways that protect the privacy of user data, often originating from as-a-service platforms.
This interdisciplinary field spans Artificial Intelligence (AI), statistical methods (such as Differential Privacy (DP) and k-Anonymity), and cryptographic techniques (such as Homomorphic Encryption (HE) and Multi-Party Computation (MPC)). Moreover, entirely new AI paradigms, such as Federated Learning (FL), have emerged specifically to address privacy concerns. The integration of these tools, and particularly the intersection of their applications, promises to drive the next generation of PP-MDL, which is anticipated to become the standard in the coming years. Despite recent exponential growth in PP-MDL literature, a single, definitive solution remains elusive, as each technique has distinct advantages and limitations. These challenges are actively addressed by the PP-MDL community, leveraging expertise from AI, statistics, cryptography, and computational intelligence.
Overcoming these challenges is essential for advancing PP-MDL, enabling broader adoption and fostering more powerful, privacy-aware, and secure AI solutions, thereby meeting the increasing demands from users, providers, and regulatory bodies. This Special Session aims to bring together innovative contributions that can drive the development of more effective and efficient PP-MDL solutions. The diversity of attendees at IJCNN presents a unique opportunity to broaden the impact of this field, welcoming fresh perspectives and exploring new research directions, application scenarios, and interdisciplinary contributions.
Key Information
- Date: June 30 - July 5, 2025
- Location: Rome, Italy
- Submission Deadline:
15 January 202530 January 2025
Call for Papers
This Special Session invites papers on the following topics:
- Design of new families of ML and DL models for PP-MDL
- Use of HE for encrypted inference and training of PP-MDL models
- Advances in MPC for privacy-preserving ML
- Applications of DP and k-Anonymity in PP-MDL
- Privacy-Preserving FL approaches
- Hardware accelerators for PP-MDL
- Ethical considerations in PP-MDL
- Innovative application scenarios for PP-MDL
- PP-MDL for genomic and health data
- Software frameworks supporting PP-MDL
- PP-MDL as-a-service solutions
Submit your paper in the Main Track and select this special session as primary Subject Area.
Contact
If you have any questions, please contact:
