Nearly query-optimal classical shadow estimation of unitary channels

Abstract: Classical shadow estimation (CSE) is a powerful tool for learning properties of quantum states and quantum processes. Here we consider the CSE task for quantum unitary channels. By querying an unknown unitary channel $\mathcal{U}$ multiple times in quantum experiments, the goal is to learn a classical description of $\mathcal{U}$ such that one can later use it to accurately predict many different linear properties of the channel, i.e., the expectation values of arbitrary observables measured on the output of $\mathcal{U}$ upon arbitrary input states. Based on collective measurements on multiple systems, we propose a query efficient protocol for this task, whose query complexity achieves a quadratic advantage over previous best approach for this problem, and almost saturates the information-theoretic lower bound. To enhance practicality, we also present a variant protocol using only single-copy measurements, which still offers better query performance than any previous protocols that do not use additional quantum memories. In addition to linear properties, our protocol can also be applied to simultaneously predict many non-linear properties such as out-of-time-ordered correlators. Given the importance of CSE, this work may represent a significant advance in the study of learning unitary channels.