A polynomial-time dissipation-based quantum algorithm for solving the ground states of a class of classically hard Hamiltonians
Zhong-Xia Shang,Zi-Han Chen,Chao-Yang Lu,Jian-Wei Pan,Ming-Cheng Chen
2024-07-04
Abstract:In this work, we give a quantum algorithm for solving the ground states of a class of Hamiltonians. The mechanism of the exponential speedup that appeared in our algorithm comes from dissipation in open quantum systems. To utilize the dissipation, the central idea is to treat $n$-qubit density matrices $\rho$ as $2n$-qubit pure states $|\rho\rangle$ by vectorization and normalization. By doing so, the Lindblad master equation (LME) becomes a Schrödinger equation with non-Hermitian Hamiltonian $L$. The steady-state $\rho_{ss}$ of the LME, therefore, corresponds to the ground states $|\rho_{ss}\rangle$ of Hamiltonians with the form $L^†L$. The runtime of the LME has no dependence on $\zeta$ the overlap between the initial state and the ground state compared with the Heisenberg scaling $\mathcal{O}(\zeta^{-1})$ in other algorithms. For the input part, given a Hamiltonian $H$, under plausible assumptions, we give a polynomial-time classical procedure to judge and solve whether there exists $L$ such that $H-E_0=L^†L$. For the output part, we define the mission as estimating expectation values of arbitrary operators with respect to the ground state $|\rho_{ss}\rangle$, which can be done surprisingly by an efficient measurement protocol on $\rho_{ss}$ with no need to prepare $|\rho_{ss}\rangle$. We give several pieces of evidence on the quantum hardness of really preparing $|\rho_{ss}\rangle$, which indicates a potential complexity separation between our algorithm and those projection-based quantum algorithms such as quantum phase estimation. Further, we show that the Hamiltonians that can be efficiently solved by our algorithms contain classically hard instances assuming $\text{P}\neq \text{BQP}$. Later, we discuss and analyze several important aspects of the algorithm including generalizing to other types of Hamiltonians and the "non-linear`` dynamics in the algorithm.
Quantum Physics