Determining system Hamiltonian from eigenstate measurements without correlation functions

NEW JOURNAL OF PHYSICS(2019)

引用 8|浏览3
暂无评分
摘要
For a local Hamiltonian $H=\sum_i c_i A_i$, with $A_i$s being local operators, it is known that $H$ could be encoded in a single (non-degenerate) eigenstate $|\psi\rangle$ in certain cases. One case is that the system satisfies the Eigenstate Thermalization Hypothesis (ETH), where the local reduced density matrix asymptotically become equal to the thermal reduced density matrix [PRX \textbf{8}, 021026 (2018)]. In this case, one can reproduce $H$ (i.e. $c_i$s) from local measurement results $\langle\psi|A_i|\psi\rangle=a_i$. Another case is that the two-point correlation functions $\langle\psi|A_iA_j|\psi\rangle$ are known, one can reproduce $H$ without satisfying ETH (arXiv: 1712.01850); however, in practice nonlocal correlation functions $\langle\psi|A_iA_j|\psi\rangle$ are not easy to obtain. In this work, we develop a method to determine $H$ (i.e., $c_i$s) with local measurement results $a_i=\langle\psi|A_i|\psi\rangle$ and without the ETH assumption, by reformulating the task as an unconstrained optimization problem of certain target function of $c_i$s, with only polynomial number of parameters in terms of system size when $A_i$s are local operators. Our method applies in general cases for the known form of $A_i$s, and is tested numerically for both randomly generated $A_is$ and also the case when $A_i$s are local operators. Our result shed light on the fundamental question of how a single eigenstate can encode the full system Hamiltonian, indicating a somewhat surprising answer that only local measurements are enough without additional assumptions, for generic cases.
更多
查看译文
关键词
eigenstate, Hamiltonian determination, gradient based optimization method, no correlation functions
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要