AI RESEARCH
Optimal differentially private kernel learning with random projection
arXiv CS.LG
•
ArXi:2507.17544v4 Announce Type: replace-cross Differential privacy has become a cornerstone in the development of privacy-preserving learning algorithms. This work addresses optimizing differentially private kernel learning within the empirical risk minimization (ERM) framework. We propose a novel differentially private kernel ERM algorithm based on random projection in the reproducing kernel Hilbert space using Gaussian processes.