Thermodynamic interpretation to Stochastic Fisher Information and Single-Trajectory Speed Limits

Pedro B. Melo, Fernando Iemini, Diogo O. Soares-Pinto, Sílvio M. Duarte Queirós, Welles A. M. Morgado · 2025

Read Thermodynamic interpretation to Stochastic Fisher Information and Single-Trajectory Speed Limits by Pedro B. Melo, Fernando Iemini, Diogo O. Soares-Pinto, Sílvio M. Duarte Queirós, Welles A. M. Morgado (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-20890.txt.

Introduction

Pedro Melo and co-authors explored a measure of change in random systems in two thousand twenty-five. The measure, called stochastic Fisher information, tracks single paths a system takes. On average it matches a classic measure used in heat science. That classic tool helps define lengths and distances from balance states. What the single-path version means for heat and work was unclear.

The authors show its meaning rests on two things only. One is the disorder the system produces along a path. The other is the force driving it out of balance. From it they derive speed limits for a single path. Those limits cap how fast one run of a process can evolve. The work tightens the math behind tiny engines and sensors.

Catalog record

CREATORPedro B. Melo, Fernando Iemini, Diogo O. Soares-Pinto, Sílvio M. Duarte Queirós, Welles A. M. Morgado
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE45 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-20890

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Thermodynamic interpretation to Stochastic Fisher Information and Single-Trajectory Speed Limits Pedro B. Melo, Fernando Iemini, Diogo O. Soares-Pinto, Sílvio M. Duarte Queirós, Welles A. M. Morgado Abstract The Fisher information (FI) metric is a Riemannian metric that allows a geometric treatment of stochastic thermodynamics, introducing the possibility of computing thermodynamic lengths and deviations from equilibrium. At the trajectory level, a related quantity can be introduced, the stochastic Fisher information (SFI), which on average, is equivalent to the FI. In this work, we discuss two fundamental questions regarding the SFI; namely, (i) what is the thermodynamic interpretation of the SFI is and (ii) whether there are any trajectory-level thermodynamic bounds due to the SFI. We find that, contrary to previous results in the literature for the FI, the thermodynamic interpretation of the SFI depends only on the entropy produced by the system and on the thermodynamic force. Moreover, we find that the SFI allows one to derive single-trajectory speed limits, which we demonstrate to hold for a Brownian particle under a saturating drive force and a Brownian particle under a decreasing drive force. From the ensemble of single-trajectory bounds one can derive a hierarchy of average speed limits that are always less tight than the one derived from the FI. We test our results for speed limits on the adopted models and find that the hierarchy of average speed limits is respected and that the single-trajectory speed limits behave qualitatively similarly to the average and stochastic speed limits, with some trajectories achieving velocities higher than the tightest average bound whenever it does not saturate. Our results open avenues for the exploration of uncertainty relations at the trajectory level. I Introduction The roots of Stochastic Thermodynamics stem from the introduction of fluctuation theorems (FTs) Evans et al. 1993 ; Gallavotti and Cohen 1995a ; Gallavotti and Cohen 1995b ; Jarzynski 1997 ; Crooks 1999 ; Martins et al. 2025 . In essence, FTs show that the laws of thermodynamics for non-equilibrium regimes are probabilistic instead of deterministic and that for systems significantly out of the macroscopic limit the distributions of all forms of energy (including heat and work), and entropy (namely lack of information) have nonzero cumulants beyond first order Sevick et al. 2008 ; Seifert 2012 ; Landi and Paternostro 2021 ; Martins et al. 2025 . This probabilistic framework is called Stochastic Thermodynamics and in this context one can also formulate a set of laws analogous to the equilibrium thermodynamic laws Sekimoto 2010 ; Peliti and Pigolotti 2021 . For the second law, it becomes possible to calculate stochastic entropy and stochastic entropy production along a single trajectory Seifert 2005 ; Landi and Paternostro 2021 , with its probability distributions following a fluctuation theorem. These results are experimentally accessible for systems such as optical tweezers Collin et al. 2005 ; Ciliberto 2017 or even in the quantum regime Batalhão et al. 2014 ; Micadei et al. 2019 . Figure 1: Schematic representation of the similarities between the SFI taking \theta=t . The upper boxes indicate the FI and the SFI frameworks, with their corresponding definitions along, and the symbol that represents the respective length in each framework. The arrow between upper balloons indicates that the FI can be obtained by averaging the SFI. The lower boxes represent the geometric relationships for both FI and SFI, for the respective lengths, action along the curve for each metric, and each respective uncertainty relation. The dashed lines between the lower ballons indicate the relations between the thermodynamic length \mathcal{L}(t) and the average stochastic length \langle l[x(t)]\rangle , as well as between the thermodynamic action \mathcal{J}(t) and the average stochastic action \langle j[x(t)]\rangle . A geometrical approach to thermodynamics is also meaningful and has first been established for a slow driving regime in usual thermodynamics Weinhold 1975 ; Ruppeiner 1979 ; Salamon and Berry 1983 ; Brody and Rivier 1995 ; Ruppeiner 1995 . More recently, an approach valid for more general non-equilibrium regimes have been established, and extended to stochastic systems as well Crooks 2007 ; Feng and Crooks 2009 ; Sivak and Crooks 2012 ; Ito 2018 ; Ito and Dechant 2020 . Central to this approach is information geometry, a framework that introduces statistical manifolds, i.e. Riemannian manifolds with probability distributions as points in state space Amari and Nagaoka 2000 . By the Čencov’s theorem one learns that the only metric for the probability manifold that is contractive under noisy transformations is the Fisher information (FI) metric Fisher 1922 ; Fisher 1925 . The FI is a quantity that has been multidisciplinarily employed in areas such as quantum information Escher et al. 2011 ; Tóth 2012 ; Pires et al. 2016 ; Liu et al. 2019 ; Yu et al. 2022 ; Yadin et al. 2024 ; Iemini et al. 2024 ; Bettmann and Goold 2025 , ecology Mayer et al. 2006 , and finance Aït-Sahalia and Jacod 2008 , to mention a few. This quantity can be interpreted as a way to measure the amount of information a random variable X has about an unknown parameter \theta . This interpretation is synthesized in the form of the Cramér-Rao (CR) bound Rao 1992 . It states that for an unknown parameter \theta , with an estimator \hat{\Theta}(\theta) (for instance, a sample mean of a probability distribution of \theta within a data set), the precision of \hat{\Theta}(\theta) is at most the FI. The FI and the CR bound are directly related to thermodynamic inequalities, such as the thermodynamic uncertainty relations (TURs) Barato and Seifert 2015 ; Salazar 2022 ; Van Vu and Saito 2023 , the kinetic uncertainty relations (KURs) Dechant and ichi Sasa 2020 , and the stochastic speed limits Shiraishi et al. 2018 ; Gupta and Busiello 2020 ; Vo et al. 2020 . They relate a signal-to-noise ratio (in the case of TURs) and the speed of transformation of states (in the case of speed limits) to the entropy production rate. In the quantum regime, there is known to exist a quantum FI that measures the amount of information a quantum state has about an unknown parameter \theta . The QFI is also directly related to (quantum) speed limits Taddei et al. 2013 ; del Campo et al. 2013 ; Pires et al. 2016 ; Deffner and Campbell 2017 ; Nishiyama and Hasegawa 2025 , that roughly speaking are derived from time-energy uncertainty relation. The stochastic speed limits (obtained by FI metrics) have an analogous interpretation to quantum speed limits Okuyama and Ohzeki 2018 ; Shanahan et al. 2018 . Near-equilibrium, the stochastic speed limits affirm that the time taken for a system to go from an initial state to a final state is lower bounded by the inverse of variance of energy Ito 2018 . Recently, a stochastic quantity analogous to FI was introduced, dubbed stochastic Fisher information (SFI). For a parameter \theta , the SFI is a random variable with a probability distribution whose average reproduces the FI of the same \theta Melo et al. 2025 . The SFI distribution was also discovered to be subject to fluctuation theorems Melo et al. 2024 . As a metric, the length given by the SFI is called stochastic length

Read the complete text