Current students


FALCETTA ALESSANDROCycle: XXXVII

Section: Computer Science and Engineering
Advisor: ROVERI MANUEL
Tutor: MARTINENGHI DAVIDE

Major Research topic:
Privacy-preserving Deep Learning: algorithms and technologies

Abstract:
Deep-learning-as-a-service is a novel and promising computing paradigm aiming at providing machine/deep learning solutions through Cloud-based computing infrastructures. Unfortunately, this approach requires to send information to be processed to the Cloud, hence having catastrophic impacts on the privacy of users.
This research project aims at defining solutions and methodologies to allow the design of privacy-preserving machine and deep learning as-a-service able to preserve the user privacy. In particular, the project will aim at defining deep learning neural networks that can be trained directly on encrypted data, thanks to Homomorphic Encryption. The possible uses are, but are not limited to, time-series forecasting, image-recognition, and anomaly detection.
The interest in privacy-preserving machine and deep learning is steadily growing, even though solutions for training deep learning neural networks directly on encrypted data are still missing. This represents a major point to be addressed in the future of privacy-preserving deep learning; this research project will focus on it from several points of view.