Recovery of a generic local Hamiltonian from a degenerate steady state

Abstract

Hamiltonian Learning (HL) is essential for validating quantum systems in quantum computing. Not all Hamiltonians can be uniquely recovered from a steady state. HL success depends on the Hamiltonian model and the steady state. Here, we analyze HL for a specific type of steady state composed of eigenstates with degenerate mixing weight, making these Hamiltonian's eigenstates indistinguishable. To overcome this challenge, we utilize the orthogonality relationship between the eigenstate space and its complement space, constructing the orthogonal space equation. By counting the number of linearly independent equations derived from a steady state, we determine the recoverability of a generic local Hamiltonian. Our scheme is applicable for generic local Hamiltonians under various steady state, therefore offering a way of measuring the degree to which a steady state characterizes a Hamiltonian.