IQUIST Young Researchers Seminar: "High quantum local differential privacy breaks entanglement", Sujeet Bhalerao, Leditzky Group

Abstract: Differential privacy provides a mathematical framework for guaranteeing privacy for sensitive data. In quantum information processing, it is important to understand when privacy requirements are compatible with the preservation of quantum resources such as entanglement. We study this question for quantum local differential privacy. Our main result shows that sufficiently private quantum channels are necessarily entanglement-breaking, meaning that they destroy any entanglement between their input and a reference system. We also prove an approximate version of this result and study its consequences for private quantum learning, obtaining lower bounds on the number of samples needed for certain learning tasks even when the learner has access to quantum memory. This is joint work with Theshani Nuradha and Felix Leditzky. 

Bio: Sujeet Bhalerao is a mathematics PhD student working with Felix Leditzky on quantum information theory.