Many quantum information processing tasks spend non-negligible computational costs for preparing an input quantum state. However, a quantum input state prepared for a specific algorithm cannot be reused for another task once measured by the postulate of the quantum measurement. Moreover, the quantum state cannot be cloned. Hence, in general, one is forced to repeat the state preparation routine per algorithm, even when individual algorithms receive the same input. Meanwhile, many quantum algorithms demand repetitions for sampling the answer. Thus while information processing tasks have the potential to benefit from laws of quantum mechanics, they also impose unavoidable redundancy. Here we introduce quantum information forking that allows an array of qubits to undergo independent processes in superposition to reduce the number of the state initialization procedure. As an example, we demonstrate the application of quantum forking to quantum Monte-Carlo sampling. In this case, quantum forking allows for the implementation of the independent propagation of the quantum trajectories, while maintaining the constant cost of initial state preparation.
This work was supported in part by the Ministry of Science and ICT, Korea, under an ITRC Program, IITP-2018-2018-0-01402.