no code implementations • 7 Nov 2022 • Haleh Hayati, Carlos Murguia, Nathan van de Wouw
We present a framework for designing distorting mechanisms that allow remotely operating anomaly detectors while preserving privacy.
no code implementations • 5 Apr 2022 • Haleh Hayati, Carlos Murguia, Nathan van de Wouw
The idea is to immerse the learning algorithm, a Stochastic Gradient Decent (SGD), into a higher-dimensional system (the so-called target system) and design the dynamics of the target system so that: the trajectories of the original SGD are immersed/embedded in its trajectories, and it learns on encrypted data (here we use random matrix encryption).
no code implementations • 30 Nov 2021 • Haleh Hayati, Carlos Murguia, Nathan van de Wouw
We formulate the synthesis of distorting mechanisms in terms of semidefinite programs in which we seek to minimize the mutual information (our privacy metric) between private data and the disclosed distorted data given a desired distortion level -- how different actual and distorted data are allowed to be.