Proving Obliviousness of Probabilistic Algorithms with Formal Verification.

Obliviousness is a security feature to protect sensitive information from an algorithm’s observable behaviours. For better run-time performance, many oblivious algorithms published recently are probabilistic instead of deterministic, but this also brings difficulties to reason about them. We introduce some challenges, works and further ideas in this paper, including the concrete contribution of extending PSL for verifying one of the random oblivious algorithms. We also present my further PhD research plan on this topic.