sde_srk3_step

Contents

sde_srk3_step#

class braintools.quad.sde_srk3_step(df, dg, y, t, *args, **kwargs)#

Stochastic Runge–Kutta 3 (Stratonovich; Heun-RK3).

Uses a classic 3-stage RK3 scheme on the Stratonovich increment F(y,t) = f(y,t) dt + g(y,t) dW with a single shared dW:

  • k1 = F(y, t)

  • k2 = F(y + 0.5 k1, t + 0.5 dt)

  • k3 = F(y - k1 + 2 k2, t + dt)

  • y_{n+1} = y + (k1 + 4 k2 + k3) / 6

Parameters:

  • df (Callable[[PyTree, float | Quantity, ...], PyTree]) – As usual; Stratonovich interpretation assumed.

  • dg (Callable[[PyTree, float | Quantity, ...], PyTree]) – As usual; Stratonovich interpretation assumed.

  • y (PyTree) – As usual; Stratonovich interpretation assumed.

  • t (float | Quantity) – As usual; Stratonovich interpretation assumed.

  • *args – As usual; Stratonovich interpretation assumed.

Returns:

The updated state y_{n+1}.

Return type:

PyTree

Contents