Microscopic reweighting for nonequilibrium steady-state dynamics

Area of Science:

• Computational physics
• Statistical mechanics
• Chemical kinetics

Background:

• Computer simulations typically yield data at a single thermodynamic state.
• Statistical reweighting is established for equilibrium properties.
• Extending these methods to nonequilibrium states is a significant challenge.

Purpose of the Study:

• To extend statistical reweighting techniques to nonequilibrium steady states.
• To develop a method for reweighting dynamical properties out of equilibrium.
• To identify invariant quantities for dynamical reweighting.

Main Methods:

• Utilizing a maximum path entropy formalism under physical constraints.
• Applying stochastic thermodynamics to relate forward and backward pathway probabilities.
• Constructing pathways through Markovian transitions to avoid combinatorial issues.

Main Results:

• Developed an analytical framework for reweighting properties in and out of equilibrium.
• Successfully avoided the combinatorial explosion of microtrajectories.
• Identified a quantity invariant to dynamical reweighting, analogous to equilibrium density of states.

Conclusions:

• The proposed method enables efficient reweighting of dynamical properties in nonequilibrium systems.
• This work provides a powerful tool for analyzing complex molecular dynamics.
• The identified invariant quantity offers new insights into nonequilibrium statistical mechanics.