SMD-PSO

class globe.optimizers.particles.common_noise.SMD.SMD_PSO.SMD_PSO(bounds, n_particles=200, iter=1000, dt=0.1, beta=100000.0, alpha=1, gamma=1, lambda_=1e-10, delta=2.1, moment='M1', filter_type=None, verbose=False)[source]

Bases: SMD_Optimizer

Interface for the Stochastic Moment Dynamics PSO optimizer with optional particle filtering.

Parameters:

  • bounds (ndarray) – The bounds of the search space.

  • n_particles (int) – The number of particles.

  • iter (int) – The number of iterations.

  • dt (float) – The time step.

  • beta (float) – The inverse temperature.

  • alpha (float) – The coefficient to decrease the step size.

  • filter_type (str or None, optional) – The type of filter to apply to particles: - None: No filtering (default) - “quantile”: Filters out particles judged as non-relevant based on quantile

  • gamma (float) – The coefficient for the common noise.

  • lambda_ (float) – The regularization parameter for the common noise.

  • delta (float) – The parameter for the Bessel process.

  • moment (str) – The type of moment used for the common noise (“M1” | “M2” | “VAR” | “MVAR”).

  • verbose (bool) – Whether to print information about the optimization process.