IQIM Postdoctoral and Graduate Student Seminar
Abstract: With the development of cloud quantum computing, clients will want to ensure confidentiality of their data and algorithms (blindness), and the integrity of their computations (verifiability). In this talk, I present our recent work on two directions of research related to blind and verified quantum computing.
1.) While protocols for blind and verifiable quantum computation exist, they suffer from high overheads and from over-sensitivity: When running on noisy devices, imperfections trigger the same detection mechanisms as malicious attacks, resulting in perpetually aborted computations. We introduce the first blind and verifiable protocol for delegating BQP computations to a powerful server with repetitions as the only overhead. It is composable and statistically secure with exponentially low bounds and can tolerate a constant amount of global noise.
2.) Existing protocols for verified quantum computing fall short of resolving conflicts between client and server from a third party's perspective. In case the client rejects the outcome of a delegated computation on the basis of failed verification, a third party will generally not be able to tell whether the abort of the protocol has been triggered by deviations by the client or by the server. We propose a new cryptographic protocol that provides the same security guarantees as other blind verification protocols, while additionally resolving this conflict.