Improvement of quantum walk-based search algorithms in single marked vertex graphs
Quantum walks are powerful tools for building quantum search algorithms or quantum sampling algorithms named the construction of quantum stationary state. However, the success probability of those algorithms are all far away from 1. Amplitude amplification is usually used to amplify success probability, but the soufflé problems follow. Only stop at the right step can we achieve a maximum success probability. Otherwise, as the number of steps increases, the success probability may decrease, which will cause troubles in practical application of the algorithm when the optimal number of steps is not known. In this work, we define generalized interpolated quantum walks, which can both improve the success probability of search algorithms and avoid the soufflé problems. Then we combine generalized interpolation quantum walks with quantum fast-forwarding. The combination both reduce the times of calling walk operator of searching algorithm from $Θ((\varepsilon^{-1})\sqrt{\Heg})$ to $Θ(\log(\varepsilon^{-1})\sqrt{\Heg})$ and reduces the number of ancilla qubits required from $Θ(\log(\varepsilon^{-1})+\log\sqrt{\Heg})$ to $Θ(\log\log(\varepsilon^{-1})+\log\sqrt{\Heg})$, and the souffle problem is avoided while the success probability is improved, where $\varepsilon$ denotes the precision and $\Heg$ denotes the classical hitting time. Besides, we show that our generalized interpolated quantum walks can be used to improve the construction of quantum states corresponding to stationary distributions as well. Finally, we give an application that can be used to construct a slowly evolving Markov chain sequence by applying generalized interpolated quantum walks, which is the necessary premise in adiabatic stationary state preparation.