Abstract:It is of great interest to understand the thermalization of open quantum many-body systems, and how quantum computers are able to efficiently simulate that process. A recently introduced disispative evolution, inspired by existing models of open system thermalization, has been shown to be efficiently implementable on a quantum computer. Here, we prove that, at high enough temperatures, this evolution reaches the Gibbs state in time scaling logarithmically with system size. The result holds for Hamiltonians that satisfy the Lieb-Robinson bound, such as local Hamiltonians on a lattice, and includes long-range systems. To the best of our knowledge, these are the first results rigorously establishing the rapid mixing property of high-temperature quantum Gibbs samplers, which is known to give the fastest possible speed for thermalization in the many-body setting. We then employ our result to the problem of estimating partition functions at high temperature, showing an improved performance over previous classical and quantum algorithms.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to efficiently prepare quantum Gibbs states at high temperatures. Specifically, the authors focus on how to use quantum computers to efficiently simulate the thermalization process of open quantum many - body systems and prove that at a sufficiently high temperature, this evolution can reach the Gibbs state within a short time related to the logarithm of the system size. This involves the strict establishment of the rapid - mixing property of high - temperature quantum Gibbs samplers, which is known to be the fastest possible speed to achieve thermalization in a many - body setting. ### Main contributions 1. **Proof of rapid - mixing property**: The authors prove that for systems satisfying the Lieb - Robinson bound (such as local Hamiltonians), above a certain critical temperature \(\beta^*\), the evolution process can quickly reach the Gibbs state. This result is applicable not only to short - range interacting systems but also to long - range interacting systems. 2. **Efficient quantum algorithm**: Based on the rapid - mixing property, the authors show how to efficiently prepare Gibbs states on a quantum computer, with the required time being \(eO(n)\), where \(n\) is the system size. 3. **Efficient estimation of partition functions**: As an application, the authors show how to use the Gibbs sampling algorithm as a subroutine to efficiently estimate quantum partition functions at high temperatures. For Hamiltonians with a finite range, their method has a polynomial speedup compared to the classical method based on cluster expansion. For long - range interacting systems, to the authors' knowledge, this is the first efficient quantum partition function estimation method. ### Formulas - **Rapid - mixing property**: \[ \left\| e^{tL(\beta)}(\rho) - \sigma_\beta \right\|_1 \leq \epsilon \quad \text{for all} \quad t = \Omega(\log(n/\epsilon)) \] where \(\left\| \cdot \right\|_1\) represents the trace norm and \(\sigma_\beta\) is the Gibbs state. - **Oscillator norm**: \[ |||X||| := \sum_{a \in \Lambda} \left\| \delta_a(X) \right\|_\infty \] where \(\delta_a(X) := X - \frac{1}{2} I_a \otimes \text{tr}_a(X)\). - **Local condition**: \[ \sum_{a} \sum_{b \neq a} \kappa_{a,b} + \sum_{a} \gamma_a \leq \kappa \] where \(\kappa_{a,b}\) and \(\gamma_a\) are positive constants, satisfying: \[ \left\| [\delta_a, L_b^\dagger(\beta)](X) \right\|_\infty \leq \sum_c \kappa_{a,b} \left\| \delta_c(X) \right\|_\infty \] \[ \left\| \delta_a (L_a^\dagger(\beta) - L_a^\dagger(0))(X) \right\|_\infty \leq \sum_c \gamma_a \left\| \delta_c(X) \right\|_\infty \] ### Comparison with previous work - **Exponential - time convergence**: Until recently, quantum algorithms for preparing Gibbs states of general Hamiltonians could only guarantee convergence in exponential time. - **Polynomial - time algorithms**: Polynomial - time algorithms were limited to one - dimensional systems or commuting Hamiltonians. - **Efficiency**: The authors' method provides a quasi - linear preparation time in the high - temperature range, while the previous best result was quasi - quadratic. Through these contributions, this paper significantly advances our understanding of the complexity of high - temperature quantum Gibbs states.

Optimal quantum algorithm for Gibbs state preparation

Efficient thermalization and universal quantum computing with quantum Gibbs samplers

Quantum Thermal State Preparation

Quantum computational advantage with constant-temperature Gibbs sampling

Gibbs Sampling gives Quantum Advantage at Constant Temperatures with O(1)-Local Hamiltonians

Mixing time of quantum Gibbs sampling for random sparse Hamiltonians

Many-body thermal states on a quantum computer: a variational approach

Slow Mixing of Quantum Gibbs Samplers

An efficient and exact noncommutative quantum Gibbs sampler

Dissipative Quantum Gibbs Sampling

Quasi-optimal sampling from Gibbs states via non-commutative optimal transport metrics

Learning quantum Hamiltonians from high-temperature Gibbs states and real-time evolutions

Quantum Algorithm to Prepare Quasi-Stationary States

Robust Extraction of Thermal Observables from State Sampling and Real-Time Dynamics on Quantum Computers

Gibbs state sampling via cluster expansions

Efficient learning of ground and thermal states within phases of matter

Gibbs state preparation for commuting Hamiltonian: Mapping to classical Gibbs sampling

Quantum thermalization of translation-invariant systems at high temperature

Variational preparation of finite-temperature states on a quantum computer

Variational Gibbs State Preparation on NISQ devices