SciML / StochasticDiffEq.jl

Solvers for stochastic differential equations which connect with the scientific machine learning (SciML) ecosystem
Other
245 stars 66 forks source link

add RandomTamedEM() solver for RODEs #501

Closed rmsrosa closed 1 year ago

rmsrosa commented 1 year ago

This implements the RODE equivalent of the "Tamed Euler" SDE solver proposed in Strong convergence of an explicit numerical method for SDEs with nonglobally Lipschitz continuous coefficients by Martin Hutzenthaler, Arnulf Jentzen, Peter E. Kloeden.

This scheme, for a RODE $dX_t/dt = f(t, X_t, Y_t)$, with noise $Yt$, is an explicit scheme consisting of the time steps $$X{t{j+1}}^n = X{t_j}^n + \frac{\Delta t f(tj, X{tj}, Y{t_j})}{1 + \Delta t ||f(tj, X{tj}, Y{t_j})||}$$