Secure Sparse Matrix Multiplications and their Applications to Privacy-Preserving Machine Learning
Abstract
To preserve data privacy, multi-party computation (MPC) enables executing Machine Learning (ML) algorithms on private data. However, MPC frameworks do not include optimized operations on sparse data. This absence makes them unsuitable for ML applications involving sparse data; e.g., recommender systems or genomics. Even in plaintext, such applications involve high-dimensional sparse data, that cannot be processed without sparsity-related optimizations due to prohibitively large memory requirements.
Fields of Study & Subjects
Random Matrices and Applications
Cryptography and Data Security
Mathematical Analysis and Transform Methods
Publication details
- Journal / source
- Proceedings of the Sixteenth ACM Conference on Data and Application Security and Privacy
- Publisher
- ACM
- Pages
- 71–84
- Publication date
- 12 June 2026
- Language
- EN