arXiv:2610.00246
Physical-Work Fluctuation Relations from Accessible Quantum Macrostates
Borhan Ahmadi
Abstract
Jarzynski's equality recovers an equilibrium free-energy difference from nonequilibrium work trajectories, but its exponential average can converge very slowly because rare trajectories carry large weight. We show that coarse thermodynamic information measured at the nonequilibrium endpoint can reduce this sampling burden while keeping the same microscopic trajectories and the same free-energy target. The endpoint mean energy and a coarse spatial record define a maximum-entropy state and select one member of an exact family of fluctuation relations. That physically selected member provides an exact statistical control for the ordinary Jarzynski estimator, even when the final energy and retained record do not commute. In a finite Bose--Hubbard system, this control substantially lowers the finite-confidence sampling cost, while a small controlled departure from exact endpoint matching produces a square-root information--sampling tradeoff.
AI Summary
Can independently measured, coarse information about a nonequilibrium quantum endpoint reduce the sampling cost of estimating the ordinary Jarzynski free-energy difference without changing the drive or TPM work? This paper constructs a maximum-entropy endpoint state from its mean energy and coarse projective-record probabilities, then uses the associated fluctuation identity as a control variate for the Jarzynski estimator. For canonical preparation and unitary driving, an exact family satisfies $\langle e^{-\sigma_\vartheta}\rangle=1$. Unitarity and cancellation of the initial Gibbs weights establish the identity, including noncommuting final energy and record measurements. For the endpoint-matching member, the mean decomposes into unresolved relative entropy, a measurement-disturbance term, and a nonnegative Golden–Thompson normalization correction. Subtracting $c(e^{-\sigma_\vartheta}-1)$ preserves the Jarzynski mean for independently fixed parameters and coefficient. Exact finite-system calculations use four bosons on four Bose–Hubbard sites, with Hilbert-space dimension 35. At 1% relative accuracy and 99% confidence, a two-sided Chernoff repetition benchmark decreases from $5.6463\times10^9$ to $3.5873\times10^9$ production runs with the endpoint-matching control. Allowing relative-entropy displacement $\delta=0.01$ lowers it to $2.9496\times10^9$. Locally, the optimized fractional improvement over endpoint matching is $1.69\sqrt{\delta}+O(\delta)$, derived from quadratic information geometry and first-order cost variation. These are model-conditioned sufficient benchmarks, not minimal or end-to-end sample complexities: endpoint calibration, pilot sampling, and auxiliary-normalization uncertainty are excluded. No general sampling advantage is established. Arbitrary initial states require additional energy-resolved preparation information; the coarse energy-and-record constraints generally do not suffice.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.