Differentially Private Transferrable Deep Learning with Membership-Mappings
This work addresses privacy-preserving transfer learning for domains like image recognition, but it appears incremental as it builds on existing concepts with specific adaptations.
The paper tackles the problem of differentially private semi-supervised transfer and multi-task learning by proposing a method that combines a novel deep autoencoder variant with a tailored noise mechanism to achieve privacy with minimal data perturbation, and experiments on datasets like MNIST and Caltech256 verify its competitive robust performance.
This paper considers the problem of differentially private semi-supervised transfer and multi-task learning. The notion of \emph{membership-mapping} has been developed using measure theory basis to learn data representation via a fuzzy membership function. An alternative conception of deep autoencoder, referred to as \emph{Conditionally Deep Membership-Mapping Autoencoder (CDMMA)}, is considered for transferrable deep learning. Under practice-oriented settings, an analytical solution for the learning of CDMMA can be derived by means of variational optimization. The paper proposes a transfer and multi-task learning approach that combines CDMMA with a tailored noise adding mechanism to achieve a given level of privacy-loss bound with the minimum perturbation of the data. Numerous experiments were carried out using MNIST, USPS, Office, and Caltech256 datasets to verify the competitive robust performance of the proposed methodology.