types.backends.NoiseType

types.backends.NoiseType()

Noise structure classification for stochastic systems.

Categories: - ‘additive’: g(x,u,t) = constant (state-independent) * Most efficient - can precompute * Example: dx = f(x)dt + σ*dW

‘multiplicative’: g(x,u,t) depends on state
State-dependent noise intensity
Example: dx = f(x)dt + σxdW (Geometric Brownian Motion)
‘diagonal’: g(x,u,t) is diagonal matrix
Independent noise sources
Enables element-wise solvers
‘scalar’: Single noise source (nw=1)
Simplest stochastic case
One Wiener process
‘general’: Full coupling, no special structure
Most general, least efficient

# types.backends.NoiseType { #cdesym.types.backends.NoiseType }

```python
types.backends.NoiseType()
```

Noise structure classification for stochastic systems.

Categories:
- 'additive': g(x,u,t) = constant (state-independent)
* Most efficient - can precompute
* Example: dx = f(x)dt + σ*dW

- 'multiplicative': g(x,u,t) depends on state
* State-dependent noise intensity
* Example: dx = f(x)dt + σ*x*dW (Geometric Brownian Motion)

- 'diagonal': g(x,u,t) is diagonal matrix
* Independent noise sources
* Enables element-wise solvers

- 'scalar': Single noise source (nw=1)
* Simplest stochastic case
* One Wiener process

- 'general': Full coupling, no special structure
* Most general, least efficient