Theoretical seminar | 07 September 2022

Dr. Gleb Fedorov
Moscow Institute of Physics and Technology
Hardware for quantum deep learning with superconducting qubits

Recent developments in building connected systems of individual physical qubits are still being impeded by their low coherence. Since quantum error correction is difficult to implement in practice, requiring a very large number of extra physical qubits, there is an on-going search of simpler algorithms that can yield quantum speed-up but are less sensitive to noise. One of the approaches is to try to use the noisy intermediate-scale deep quantum circuits filled with various single- and two-qubit gates in a classical optimization routine, which tunes the gate parameters until the quantum circuit is suitable to solve, for instance, a classification problem. In this talk we will discuss how this idea is realized in practice using superconducting artificial atoms and what problems one can face during the development of the experimental setup.