Dresden 2026 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

SYSD: Symposium SKM Dissertation Prize 2026

SYSD 1: SKM Dissertation Prize Symposium 2026

SYSD 1.5: Hauptvortrag

Montag, 9. März 2026, 11:30–12:00, HSZ/0002

On stochastic thermodynamics under incomplete information: Thermodynamic inference from Markovian events — •Jann van der Meer — Department of Physics No. 1, Graduate School of Science, Kyoto University, Kyoto 606-8502, Japan

Stochastic thermodynamics provides the framework for analyzing thermodynamic quantities along individual trajectories of small but fully observable systems. If, however, the observable dynamics is coarse-grained, i.e., does not capture all relevant degrees of freedom, the principles of stochastic thermodynamics generally cannot be applied directly. With access to some observables, preeminent results in the field (e.g. the thermodynamic uncertainty relation) often establish lower bounds on the average entropy production in terms of such observables.

However, there is no reason why attention should be limited to bounds on the averaged entropy production only. In my talk, I present the concept of a fluctuating entropy production on the coarse-grained level, which contains more information than about averages only, thereby allowing us to, e.g., estimate the strength of driving forces and localize entropy production spatially and temporally. From a theoretical point of view, these results rely on the identification of Markovian events, a systematic way of describing the dynamics and thermodynamics of a variety of possible coarse-grained dynamics in terms of semi-Markov processes.

Keywords: Stochastic thermodynamics; Thermodynamic inference; Coarse graining