A comparative study of quantum control algorithms in quantum computation

Abstract

In quantum computation, there is a need to prepare a quantum state through controlling a time dependent Hamiltonian, which is also a problem of quantum optimal control. In this project, two algorithms, Krotov and Q-learning (QL), are used to solve this problem. From the results of numerical simulations, both algorithms could give a optimal control time dependent Hamiltonian with which the qubit would reach the target quantum state. When comparing the two algorithms, it is found that the control function generated by QL is more feasible in the laboratory environment. This finding shows the potential of using machine learning techniques to solve problems in physics.