English

Conserved Quantities in Linear and Nonlinear Quantum Search

Quantum Physics 2025-08-29 v2

Abstract

In this tutorial, which contains some original results, we bridge the fields of quantum computing algorithms, conservation laws, and many-body quantum systems by examining three algorithms for searching an unordered database of size NN using a continuous-time quantum walk, which is the quantum analogue of a continuous-time random walk. The first algorithm uses a linear quantum walk, and we apply elementary calculus to show that the success probability of the algorithm reaches 1 when the jumping rate of the walk takes some critical value. We show that the expected value of its Hamiltonian H0H_0 is conserved. The second algorithm uses a nonlinear quantum walk with effective Hamiltonian H(t)=H0+λψ2H(t) = H_0 + \lambda|\psi|^2, which arises in the Gross-Pitaevskii equation describing Bose-Einstein condensates. When the interactions between the bosons are repulsive, λ>0\lambda > 0, and there exists a range of fixed jumping rates such that the success probability reaches 1 with the same asymptotic runtime of the linear algorithm, but with a larger multiplicative constant. Rather than the effective Hamiltonian, we show that the expected value of H0+12λψ2H_0 + \frac{1}{2} \lambda|\psi|^2 is conserved. The third algorithm utilizes attractive interactions, corresponding to λ<0\lambda < 0. In this case there is a time-varying critical function for the jumping rate γc(t)\gamma_c(t) that causes the success probability to reach 1 more quickly than in the other two algorithms, and we show that the expected value of H(t)/[γc(t)N]H(t)/[\gamma_c(t) N] is conserved.

Keywords

Cite

@article{arxiv.2503.06423,
  title  = {Conserved Quantities in Linear and Nonlinear Quantum Search},
  author = {David A. Meyer and Thomas G. Wong},
  journal= {arXiv preprint arXiv:2503.06423},
  year   = {2025}
}

Comments

17 pages, 4 figures

R2 v1 2026-06-28T22:12:33.016Z