Identification and quantification of irreversibility in stochastic systems

Area of Science:

• Thermodynamics
• Statistical Mechanics
• Nanotechnology
• Biophysics

Background:

• Small, fluctuating systems like biological motors and synthetic nanomachines operate via irreversible, dissipative processes.
• These systems produce entropy, making its quantification crucial for understanding physical limits and design principles.
• Advances in single-molecule measurements and active-matter control drive progress in this field.

Purpose of the Study:

• To survey principled routes for characterizing and quantifying entropy production (EP) from time-series and trajectory data.
• To discuss methods for inferring dissipation from incomplete experimental information.
• To address how coarse-graining biases EP estimates in nanoscale systems.

Main Methods:

• Review of established and emerging techniques for estimating EP from experimental data.
• Analysis of methods for inferring dissipation from partial or coarse-grained observables.
• Discussion of theoretical frameworks for bounding EP in nonequilibrium systems.

Main Results:

• Principled routes exist for quantifying EP from time-series and trajectory data.
• Incomplete information and coarse-graining introduce systematic biases in EP estimates.
• Current methods provide a toolkit for estimating EP, but challenges remain.

Conclusions:

• Accurate quantification of EP is central to understanding nanoscale living and engineered systems.
• Unifying inference approaches is needed to obtain reliable bounds on EP.
• Further research is required to overcome challenges in EP estimation from complex experimental data.