isoexp/mab/smab_algs.py [25:41]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
    K = len(MAB)
    rewards = np.zeros((T,))
    draws = np.zeros((T,))
    N = np.ones((K,))  # number of observations of each arm
    S = np.zeros((K,))  # sum of rewards for each arm

    for k in range(K):
        a = k
        r = MAB[a].sample()

        # update quantities
        rewards[k] = r
        draws[k] = a
        S[a] += r
        N[a] += 1

    for t in range(K, T):
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -



isoexp/mab/smab_algs.py [102:119]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
    K = len(MAB)
    rewards = np.zeros((T,))
    draws = np.zeros((T,))

    N = np.ones((K,))  # number of observations of each arm
    S = np.zeros((K,))  # sum of rewards for each arm

    for k in range(K):
        a = k
        r = MAB[a].sample()

        # update quantities
        rewards[k] = r
        draws[k] = a
        S[a] += r
        N[a] += 1

    for t in range(K, T):
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -