Title:A Transition-Path-Ensembled Kinetic Equation for Stochastic Processes

Abstract:The Fokker-Planck equation is widely adopted in studies of Markov processes found in a variety of disciplines. It can be found that the evolution of probability density function in the state-space is completely determined by the jump moments of state transition of a stochastic system. Usually, the jump moments are only a function of starting and ending states of jumps, while no further limitation was imposed on the transition paths. However, if the stochastic features of transition paths required to be explicitly accounted for, then the Fokker-Planck equation is insufficient. In this paper, by define a new transition probability function, a transition-path-ensembled kinetic equation for stochastic processes was derived. By truncated in the first two orders, the kinetic equation is similar to the Fokker-Planck equation in form, but with the drift and diffusion terms expressed as the cumulants relative to the ensemble averaged paths, rather than jump moments as in the Fokker-Planck equation.