Probabilistic Optimization Algorithms

Algorithms that use probabilistic models to guide the search.

Overview

Probabilistic optimization algorithms build and update probability distributions over the solution space, using statistical methods to explore and exploit promising regions.

Available Algorithms

• Estimation of Distribution Algorithm (EDA) - Builds probability distribution of good solutions
• Cross Entropy Method - Adaptive importance sampling
• Bayesian Optimization - Gaussian process-based optimization
• Simulated Annealing - Probabilistic acceptance criterion

Usage Example

from opt.probabilistic import EstimationOfDistribution
from opt.benchmark.functions import rastrigin

optimizer = EstimationOfDistribution(
    func=rastrigin,
    lower_bound=-5.12,
    upper_bound=5.12,
    dim=10,
    max_iter=100
)
best_solution, best_fitness = optimizer.search()

See Also

• API Reference - Complete API documentation