import numpy as np

def gaussian_pdf(x, mu, sigma):
    return np.exp(-0.5 * (x - mu) ** 2 / (2 * sigma ** 2))

def log_likelihood(params, x, y):
    log_likelihood = 0
    for i in range(len(y)):
        log_likelihood += -np.log(gaussian_pdf(x[i], params[0], params[1])) - np.log(gaussian_pdf(x[i + 1], params[0], params[1]))
    return log_likelihood

def maximize_log_likelihood(params, x, y, max_iter=1000):
    max_log_likelihood = -np.inf
    best_params = None

    for i in range(max_iter):
        new_params = [p + 0.1 * np.random.normal(0, 0.1) for p in params]
        log_likelihood = log_likelihood(new_params, x, y)

        if log_likelihood > max_log_likelihood:
            max_log_likelihood = log_likelihood
            best_params = new_params

    return best_params

# 示例數(shù)據(jù)
mu = 0
sigma = 1
x = np.array([0, 1, 2, 3])
y = np.array([0, 1, 2, 3])

# 使用貝葉斯優(yōu)化算法找到最優(yōu)參數(shù)
best_params = maximize_log_likelihood(mu, sigma, x, y)
print("最優(yōu)參數(shù)：", best_params)