arXiv:2609.27181
Emergent Prethermal Symmetries for Scalable Hamiltonian Learning
Myeongjin Shin, Junseo Lee, Iman Marvian, Yu Tong
Abstract
Learning an interacting many-body Hamiltonian at the Heisenberg limit generally requires control that preserves local information for evolution times of order $1/\varepsilon$ to attain $\varepsilon$ precision. To achieve this, existing protocols often rely on trusted many-qubit operations or increasingly rapid control pulses, creating a precision barrier when gates involve unknown multi-qubit interaction or have finite duration. We show that neither resource is necessary for geometrically local Hamiltonians. Our protocol applies only static single-qubit fields, whose strength is independent of system size and grows polylogarithmically with $1/\varepsilon$. These fields generate emergent prethermal symmetries that suppress thermalization for the entire learning experiment while retaining informative symmetry-preserving dynamics. On a $d$-dimensional lattice of $n$ qubits, this enables learning every coefficient to accuracy $\varepsilon$, with failure probability at most $δ$, using total evolution time $\mathcal{O}(\log^d(1/\varepsilon)\log^2(n/δ)/\varepsilon).$ Thus, our learning protocol attains the information-theoretically optimal \(1/\varepsilon\) dependence up to polylogarithmic factors using product-state preparation, single-qubit measurements, and non-adaptive experiments. More broadly, our results establish emergent prethermal symmetries as a resource for quantum learning, opening a route to Heisenberg-limited characterization of many-body systems where fast or trusted many-qubit control is unavailable.
AI Summary
Can many-body Hamiltonians be learned at Heisenberg-limited precision without trusted entangling gates or increasingly rapid pulses? This paper gives a protocol using static single-qubit fields, product-state preparation, single-qubit measurements, and non-adaptive experiments. It considers $H=\sum_a\lambda_aP_a$ on an $n$-qubit, $d$-dimensional lattice, with fixed interaction weight and range, known Pauli terms, and $\|\lambda\|_\infty\le1$. The protocol estimates every coefficient to additive error $\varepsilon$, jointly with probability at least $1-\delta$, using total evolution time $O(\log^d(1/\varepsilon)\log^2(n/\delta)/\varepsilon)$ and $O(\log^{d+2}(1/\varepsilon)\log^2(n/\delta))$ global experiments. Sufficient field strength is $O(\max\{\log^d(1/\varepsilon),\log(1/\varepsilon)[1+\log\log(1/\varepsilon)]^3\})$, independent of $n$. Classical post-processing is polynomial in $n$, $1/\varepsilon$, and $\log(1/\delta)$. Random signed Pauli fields, weakened at selected defect sites, induce approximate prethermal charge conservation. Defect Ramsey signals encode effective local transition energies, extracted by robust frequency estimation. Analytic continuation of the ADHH prethermal recursion and connected-cluster Taylor expansions bound spatial truncation errors and derivatives of the energy map. These establish stable coefficient recovery by a contractive, preconditioned fitting algorithm. Defects separated by $O(\log(1/\varepsilon))$ can be probed simultaneously; lattice coloring supplies the parallelization overhead. The guarantees require geometrically local qubit interactions and accurately known, spatially programmable fields. Finite available field strength still imposes a precision limit. The stated SPAM robustness covers bounded measurement-probability deviations below a constant threshold; extensions to other particle types or control settings remain proposed directions.
Generative artificial intelligence tools (ChatGPT~5 and 6) assisted the authors in deriving and refining some of the proofs and in drafting and revising portions of the manuscript.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.