Distinguished Speaker Series

We will give a high-level presentation of some recent works in Quantitative Information Flow which have brought results of interest for the wider communities of Information Theory and Quantum security. The first work is based on the notion of core-concavity, and it allows to generalize some axiomatic approach for Quantitative Information Flow to an axiomatic presentation of Information Theory for general entropy functions. Several fundamental Information Theoretical results can be proven in this framework. (Joint work with Arman Khouzani and Arthur Americo) The second contribution introduces quantum quantitative information flow, a leakage framework for quantum computation. The framework is a natural extension of gain functions and is based on Buscemi’s work on the theory of statistical comparison in quantum information, theory for which a quantum equivalent to the coriaceous theorem holds. (Joint work with Arthur Americo)

The Gaussian distribution is widely used in mechanism design for differential privacy (DP). Thanks to its sub-Gaussian tail, it significantly reduces the chance of outliers when responding to queries. However, it can only provide approximate DP. In this paper, we introduce and analyse a new distribution for use in DP that is based on the Gaussian distribution, but it has improved privacy performance. The so-called offset-symmetric Gaussian tail (OSGT) distribution is obtained through using the normalized tails of two symmetric Gaussians around zero. Consequently, it can still have sub-Gaussian tail and lend itself to analytical derivations. We analytically derive the variance of the OSGT random variable and its differential privacy metrics. Numerical results show the OSGT mechanism can offer better privacy-utility performance compared to the Gaussian and Laplace mechanisms.