Langevin equations with multiplicative noise: efficient integration

An efficient and accurate integration of stochastic (partial)
differential equations with multiplicative noise can be obtained
by separating the deterministic from the stochastic terms, the latter being
treated by sampling exactly the solution of the associated
Fokker-Planck equation. We demonstrate the computational power of
this method by applying it to most absorbing phase transitions for
which Langevin equations have been proposed. This provides precise
estimates of the associated scaling exponents, clarifying the
classification of these nonequilibrium problems, and confirms or
refutes some existing theories.