Langevin dynamics for vector variables driven by multiplicative white noise: a functional formalism

TL;DR

This paper develops a functional formalism for multi-dimensional Langevin processes with multiplicative noise, simplifying the analysis of time reversal and stochastic calculus without discretization.

Contribution

It introduces a novel functional integral approach using commuting and Grassmann variables to handle stochastic processes with multiplicative noise.

Findings

Simplifies time reversal analysis in stochastic processes.

Provides a new functional representation for complex Langevin equations.

Applies the formalism to micromagnetic Landau-Lifshitz-Gilbert equations.

Abstract

We discuss general multi-dimensional stochastic processes driven by a system of Langevin equations with multiplicative white noise. In particular, we address the problem of how time reversal diffusion processes are affected by the variety of conventions available to deal with stochastic integrals. We present a functional formalism to built up the generating functional of correlation functions without any type of discretization of the Langevin equations at any intermediate step. The generating functional is characterized by a functional integration over two sets of commuting variables as well as Grassmann variables. In this representation, time reversal transformation became a linear transformation in the extended variables, simplifying in this way the complexity introduced by the mixture of prescriptions and the associated calculus rules. The stochastic calculus is codified in our formalism in the structure of the Grassmann algebra. We study some examples such as higher order derivatives Langevin equations and the functional representation of the micromagnetic stochastic Landau-Lifshitz-Gilbert equation.