    def normal_distribution(self, pos, sample):
        """returns the value of normal distribution, given the weight's sample and target position
        
        Parameters
        ----------
        pos: int
            the epoch number of the position you want to predict
        sample: list
            sample is a (1 * NUM_OF_FUNCTIONS) matrix, representing{w1, w2, ... wk}

        Returns
        -------
        float
            the value of normal distribution
        """
        curr_sigma_sq = self.sigma_sq(sample)
        delta = self.trial_history[pos - 1] - self.f_comb(pos, sample)
        return np.exp(np.square(delta) / (-2.0 * curr_sigma_sq)) / np.sqrt(2 * np.pi * np.sqrt(curr_sigma_sq))